diff --git a/README.md b/README.md
index a58cd8bf..36153746 100644
--- a/README.md
+++ b/README.md
@@ -86,3 +86,7 @@ dictionary with the following keys:
 
 and images without valid values for `rhip, rshoulder, lhip, lshoulder` are
 ignored.
+
+## Contributions
+
+- Google Colab Demo can be found here [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1nKtymlEhcE4aZs-EOTBaqPPIAKX5RQB8#scrollTo=m0TFJfYd6pOT) by Marcos Carbonell
diff --git a/demo/.gitignore b/demo/.gitignore
new file mode 100644
index 00000000..a62418ae
--- /dev/null
+++ b/demo/.gitignore
@@ -0,0 +1,9 @@
+datasets/
+__pycache__/
+log/
+data/
+*.tar.gz
+*.zip
+checkpoints/*
+
+
diff --git a/demo/README.md b/demo/README.md
new file mode 100644
index 00000000..848e8225
--- /dev/null
+++ b/demo/README.md
@@ -0,0 +1,4 @@
+# vunet-demo
+
+- This repository is a build for as a demo for the [VUnet Paper](https://arxiv.org/pdf/1804.04694.pdf) of Esser et. al. 
+- The Google Colab Demo can be found here [![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/1nKtymlEhcE4aZs-EOTBaqPPIAKX5RQB8#scrollTo=m0TFJfYd6pOT)
diff --git a/demo/arial.ttf b/demo/arial.ttf
new file mode 100644
index 00000000..886789b8
Binary files /dev/null and b/demo/arial.ttf differ
diff --git a/demo/batches_pg2.py b/demo/batches_pg2.py
new file mode 100644
index 00000000..f9396dde
--- /dev/null
+++ b/demo/batches_pg2.py
@@ -0,0 +1,408 @@
+from PIL import Image, ImageDraw, ImageFont
+import numpy as np
+import os
+import cv2
+import math
+
+def load_img(path, target_size):
+    """Load image. target_size is specified as (height, width, channels)
+    where channels == 1 means grayscale. uint8 image returned."""
+    img = Image.open(path)
+    grayscale = target_size[2] == 1
+    if grayscale:
+        if img.mode != 'L':
+            img = img.convert('L')
+    else:
+        if img.mode != 'RGB':
+            img = img.convert('RGB')
+    wh_tuple = (target_size[1], target_size[0])
+    if img.size != wh_tuple:
+        img = img.resize(wh_tuple, resample = Image.BILINEAR)
+
+    x = np.asarray(img, dtype = "uint8")
+    if len(x.shape) == 2:
+        x = np.expand_dims(x, -1)
+
+    return x
+
+
+def save_image(X, name):
+    """Save image as png."""
+    fname = os.path.join(out_dir, name + ".png")
+    Image.fromarray(X).save(fname)
+
+
+def preprocess(x):
+    """From uint8 image to [-1,1]."""
+    return np.cast[np.float32](x / 127.5 - 1.0)
+
+
+def preprocess_mask(x):
+    """From uint8 mask to [0,1]."""
+    mask = np.cast[np.float32](x / 255.0)
+    if mask.shape[-1] == 3:
+        mask = np.amax(mask, axis = -1, keepdims = True)
+    return mask
+
+
+def postprocess(x):
+    """[-1,1] to uint8."""
+    x = (x + 1.0) / 2.0
+    x = np.clip(255 * x, 0, 255)
+    x = np.cast[np.uint8](x)
+    return x
+
+def tile(X, rows, cols):
+    """Tile images for display."""
+    tiling = np.zeros((rows * X.shape[1], cols * X.shape[2], X.shape[3]), dtype = X.dtype)
+    for i in range(rows):
+        for j in range(cols):
+            idx = i * cols + j
+            if idx < X.shape[0]:
+                img = X[idx,...]
+                tiling[
+                        i*X.shape[1]:(i+1)*X.shape[1],
+                        j*X.shape[2]:(j+1)*X.shape[2],
+                        :] = img
+    return tiling
+
+
+def plot_batch(X, out_path):
+    """Save batch of images tiled."""
+    n_channels = X.shape[3]
+    if n_channels > 3:
+        X = X[:,:,:,np.random.choice(n_channels, size = 3)]
+    X = postprocess(X)
+    rc = math.sqrt(X.shape[0])
+    rows = cols = math.ceil(rc)
+    canvas = tile(X, rows, cols)
+    canvas = np.squeeze(canvas)
+    Image.fromarray(canvas).save(out_path)
+
+
+def make_joint_img(img_shape, jo, joints):
+    # three channels: left, right, center
+    scale_factor = img_shape[1] / 128
+    thickness = int(3 * scale_factor)
+    imgs = list()
+    for i in range(3):
+        imgs.append(np.zeros(img_shape[:2], dtype = "uint8"))
+
+    assert("cnose" in jo)
+    # MSCOCO
+    body = ["lhip", "lshoulder", "rshoulder", "rhip"]
+    body_pts = np.array([[joints[jo.index(part),:] for part in body]])
+    if np.min(body_pts) >= 0:
+        body_pts = np.int_(body_pts)
+        cv2.fillPoly(imgs[2], body_pts, 255)
+
+    right_lines = [
+            ("rankle", "rknee"),
+            ("rknee", "rhip"),
+            ("rhip", "rshoulder"),
+            ("rshoulder", "relbow"),
+            ("relbow", "rwrist")]
+    for line in right_lines:
+        l = [jo.index(line[0]), jo.index(line[1])]
+        if np.min(joints[l]) >= 0:
+            a = tuple(np.int_(joints[l[0]]))
+            b = tuple(np.int_(joints[l[1]]))
+            cv2.line(imgs[0], a, b, color = 255, thickness = thickness)
+
+    left_lines = [
+            ("lankle", "lknee"),
+            ("lknee", "lhip"),
+            ("lhip", "lshoulder"),
+            ("lshoulder", "lelbow"),
+            ("lelbow", "lwrist")]
+    for line in left_lines:
+        l = [jo.index(line[0]), jo.index(line[1])]
+        if np.min(joints[l]) >= 0:
+            a = tuple(np.int_(joints[l[0]]))
+            b = tuple(np.int_(joints[l[1]]))
+            cv2.line(imgs[1], a, b, color = 255, thickness = thickness)
+
+    rs = joints[jo.index("rshoulder")]
+    ls = joints[jo.index("lshoulder")]
+    cn = joints[jo.index("cnose")]
+    neck = 0.5*(rs+ls)
+    a = tuple(np.int_(neck))
+    b = tuple(np.int_(cn))
+    if np.min(a) >= 0 and np.min(b) >= 0:
+        cv2.line(imgs[0], a, b, color = 127, thickness = thickness)
+        cv2.line(imgs[1], a, b, color = 127, thickness = thickness)
+
+    cn = tuple(np.int_(cn))
+    leye = tuple(np.int_(joints[jo.index("leye")]))
+    reye = tuple(np.int_(joints[jo.index("reye")]))
+    if np.min(reye) >= 0 and np.min(leye) >= 0 and np.min(cn) >= 0:
+        cv2.line(imgs[0], cn, reye, color = 255, thickness = thickness)
+        cv2.line(imgs[1], cn, leye, color = 255, thickness = thickness)
+
+    img = np.stack(imgs, axis = -1)
+    if img_shape[-1] == 1:
+        img = np.mean(img, axis = -1)[:,:,None]
+    return img
+
+
+def valid_joints(*joints):
+    j = np.stack(joints)
+    return (j >= 0).all()
+
+
+def zoom(img, factor, center = None):
+    shape = img.shape[:2]
+    if center is None or not valid_joints(center):
+        center = np.array(shape) / 2
+    e1 = np.array([1,0])
+    e2 = np.array([0,1])
+
+    dst_center = np.array(center)
+    dst_e1 = e1 * factor
+    dst_e2 = e2 * factor
+
+    src = np.float32([center, center+e1, center+e2])
+    dst = np.float32([dst_center, dst_center+dst_e1, dst_center+dst_e2])
+    M = cv2.getAffineTransform(src, dst)
+
+    return cv2.warpAffine(img, M, shape, flags = cv2.INTER_AREA, borderMode = cv2.BORDER_REPLICATE)
+
+
+def get_crop(bpart, joints, jo, wh, o_w, o_h, ar = 1.0):
+    bpart_indices = [jo.index(b) for b in bpart]
+    part_src = np.float32(joints[bpart_indices])
+
+    # fall backs
+    if not valid_joints(part_src):
+        if bpart[0] == "lhip" and bpart[1] == "lknee":
+            bpart = ["lhip"]
+            bpart_indices = [jo.index(b) for b in bpart]
+            part_src = np.float32(joints[bpart_indices])
+        elif bpart[0] == "rhip" and bpart[1] == "rknee": 
+            bpart = ["rhip"]
+            bpart_indices = [jo.index(b) for b in bpart]
+            part_src = np.float32(joints[bpart_indices])
+        elif bpart[0] == "lshoulder" and bpart[1] == "rshoulder" and bpart[2] == "cnose": 
+            bpart = ["lshoulder", "rshoulder", "rshoulder"]
+            bpart_indices = [jo.index(b) for b in bpart]
+            part_src = np.float32(joints[bpart_indices])
+
+
+    if not valid_joints(part_src):
+            return None
+
+    if part_src.shape[0] == 1:
+        # leg fallback
+        a = part_src[0]
+        b = np.float32([a[0],o_h - 1])
+        part_src = np.float32([a,b])
+
+    if part_src.shape[0] == 4:
+        pass
+    elif part_src.shape[0] == 3:
+        # lshoulder, rshoulder, cnose
+        if bpart == ["lshoulder", "rshoulder", "rshoulder"]:
+            segment = part_src[1] - part_src[0]
+            normal = np.array([-segment[1],segment[0]])
+            if normal[1] > 0.0:
+                normal = -normal
+
+            a = part_src[0] + normal
+            b = part_src[0]
+            c = part_src[1]
+            d = part_src[1] + normal
+            part_src = np.float32([a,b,c,d])
+        else:
+            assert bpart == ["lshoulder", "rshoulder", "cnose"]
+            neck = 0.5*(part_src[0] + part_src[1])
+            neck_to_nose = part_src[2] - neck
+            part_src = np.float32([neck + 2*neck_to_nose, neck])
+
+            # segment box
+            segment = part_src[1] - part_src[0]
+            normal = np.array([-segment[1],segment[0]])
+            alpha = 1.0 / 2.0
+            a = part_src[0] + alpha*normal
+            b = part_src[0] - alpha*normal
+            c = part_src[1] - alpha*normal
+            d = part_src[1] + alpha*normal
+            #part_src = np.float32([a,b,c,d])
+            part_src = np.float32([b,c,d,a])
+    else:
+        assert part_src.shape[0] == 2
+
+        segment = part_src[1] - part_src[0]
+        normal = np.array([-segment[1],segment[0]])
+        alpha = ar / 2.0
+        a = part_src[0] + alpha*normal
+        b = part_src[0] - alpha*normal
+        c = part_src[1] - alpha*normal
+        d = part_src[1] + alpha*normal
+        part_src = np.float32([a,b,c,d])
+
+    dst = np.float32([[0.0,0.0],[0.0,1.0],[1.0,1.0],[1.0,0.0]])
+    part_dst = np.float32(wh * dst)
+
+    M = cv2.getPerspectiveTransform(part_src, part_dst)
+    return M
+
+
+def normalize(imgs, coords, stickmen, jo):
+
+    out_imgs = list()
+    out_stickmen = list()
+
+    bs = len(imgs)
+    for i in range(bs):
+        img = imgs[i]
+        joints = coords[i]
+        stickman = stickmen[i]
+
+        h,w = img.shape[:2]
+        o_h = h
+        o_w = w
+        h = h // 4
+        w = w // 4
+        wh = np.array([w,h])
+        wh = np.expand_dims(wh, 0)
+
+        bparts = [
+                ["lshoulder","lhip","rhip","rshoulder"],
+                ["lshoulder", "rshoulder", "cnose"],
+                ["lshoulder","lelbow"],
+                ["lelbow", "lwrist"],
+                ["rshoulder","relbow"],
+                ["relbow", "rwrist"],
+                ["lhip", "lknee"],
+                ["rhip", "rknee"]]
+        ar = 0.5
+
+        part_imgs = list()
+        part_stickmen = list()
+        for bpart in bparts:
+            part_img = np.zeros((h,w,3))
+            part_stickman = np.zeros((h,w,3))
+            M = get_crop(bpart, joints, jo, wh, o_w, o_h, ar)
+
+            if M is not None:
+                part_img = cv2.warpPerspective(img, M, (h,w), borderMode = cv2.BORDER_REPLICATE)
+                part_stickman = cv2.warpPerspective(stickman, M, (h,w), borderMode = cv2.BORDER_REPLICATE)
+
+            part_imgs.append(part_img)
+            part_stickmen.append(part_stickman)
+        img = np.concatenate(part_imgs, axis = 2)
+        stickman = np.concatenate(part_stickmen, axis = 2)
+
+        """
+        bpart = ["lshoulder","lhip","rhip","rshoulder"]
+        dst = np.float32([[0.0,0.0],[0.0,1.0],[1.0,1.0],[1.0,0.0]])
+        bpart_indices = [jo.index(b) for b in bpart]
+        part_src = np.float32(joints[bpart_indices])
+        part_dst = np.float32(wh * dst)
+
+        M = cv2.getPerspectiveTransform(part_src, part_dst)
+        img = cv2.warpPerspective(img, M, (h,w), borderMode = cv2.BORDER_REPLICATE)
+        stickman = cv2.warpPerspective(stickman, M, (h,w), borderMode = cv2.BORDER_REPLICATE)
+        """
+
+        """
+        # center of possible rescaling
+        c = joints[jo.index("cneck")]
+
+        # find valid body part for scale estimation
+        a = joints[jo.index("lshoulder")]
+        b = joints[jo.index("lhip")]
+        target_length = 33.0
+        if not valid_joints(a,b):
+            a = joints[jo.index("rshoulder")]
+            b = joints[jo.index("rhip")]
+            target_length = 33.0
+        if not valid_joints(a,b):
+            a = joints[jo.index("rshoulder")]
+            b = joints[jo.index("relbow")]
+            target_length = 33.0 / 2
+        if not valid_joints(a,b):
+            a = joints[jo.index("lshoulder")]
+            b = joints[jo.index("lelbow")]
+            target_length = 33.0 / 2
+        if not valid_joints(a,b):
+            a = joints[jo.index("lwrist")]
+            b = joints[jo.index("lelbow")]
+            target_length = 33.0 / 2
+        if not valid_joints(a,b):
+            a = joints[jo.index("rwrist")]
+            b = joints[jo.index("relbow")]
+            target_length = 33.0 / 2
+
+        if valid_joints(a,b):
+            body_length = np.linalg.norm(b - a)
+            factor = target_length / body_length
+            img = zoom(img, factor, center = c)
+            stickman = zoom(stickman, factor, center = c)
+        else:
+            factor = 0.25
+            img = zoom(img, factor, center = c)
+            stickman = zoom(stickman, factor, center = c)
+        """
+
+        out_imgs.append(img)
+        out_stickmen.append(stickman)
+    out_imgs = np.stack(out_imgs)
+    out_stickmen = np.stack(out_stickmen)
+    return out_imgs, out_stickmen
+
+
+def make_mask_img(img_shape, jo, joints):
+    scale_factor = img_shape[1] / 128
+    masks = 3*[None]
+    for i in range(3):
+        masks[i] = np.zeros(img_shape[:2], dtype = "uint8")
+
+    body = ["lhip", "lshoulder", "rshoulder", "rhip"]
+    body_pts = np.array([[joints[jo.index(part),:] for part in body]], dtype = np.int32)
+    cv2.fillPoly(masks[1], body_pts, 255)
+
+    head = ["lshoulder", "chead", "rshoulder"]
+    head_pts = np.array([[joints[jo.index(part),:] for part in head]], dtype = np.int32)
+    cv2.fillPoly(masks[2], head_pts, 255)
+
+    thickness = int(15 * scale_factor)
+    lines = [[
+        ("rankle", "rknee"),
+        ("rknee", "rhip"),
+        ("rhip", "lhip"),
+        ("lhip", "lknee"),
+        ("lknee", "lankle") ], [
+            ("rhip", "rshoulder"),
+            ("rshoulder", "relbow"),
+            ("relbow", "rwrist"),
+            ("rhip", "lhip"),
+            ("rshoulder", "lshoulder"),
+            ("lhip", "lshoulder"),
+            ("lshoulder", "lelbow"),
+            ("lelbow", "lwrist")], [
+                ("rshoulder", "chead"),
+                ("rshoulder", "lshoulder"),
+                ("lshoulder", "chead")]]
+    for i in range(len(lines)):
+        for j in range(len(lines[i])):
+            line = [jo.index(lines[i][j][0]), jo.index(lines[i][j][1])]
+            a = tuple(np.int_(joints[line[0]]))
+            b = tuple(np.int_(joints[line[1]]))
+            cv2.line(masks[i], a, b, color = 255, thickness = thickness)
+
+    for i in range(3):
+        r = int(11 * scale_factor)
+        if r % 2 == 0:
+            r = r + 1
+        masks[i] = cv2.GaussianBlur(masks[i], (r,r), 0)
+        maxmask = np.max(masks[i])
+        if maxmask > 0:
+            masks[i] = masks[i] / maxmask
+    mask = np.stack(masks, axis = -1)
+    mask = np.uint8(255 * mask)
+
+    return mask
+
+
+
diff --git a/demo/buffered_wrapper.py b/demo/buffered_wrapper.py
new file mode 100644
index 00000000..c97d91fb
--- /dev/null
+++ b/demo/buffered_wrapper.py
@@ -0,0 +1,23 @@
+
+from multiprocessing.pool import ThreadPool
+
+class BufferedWrapper(object):
+    """Fetch next batch asynchronuously to avoid bottleneck during GPU
+    training."""
+    def __init__(self, gen):
+        self.gen = gen
+        self.n = gen.n
+        self.pool = ThreadPool(1)
+        self._async_next()
+
+
+    def _async_next(self):
+        self.buffer_ = self.pool.apply_async(next, (self.gen,))
+
+
+    def __next__(self):
+        result = self.buffer_.get()
+        self._async_next()
+        return result
+
+
diff --git a/demo/config.py b/demo/config.py
new file mode 100644
index 00000000..fde61de2
--- /dev/null
+++ b/demo/config.py
@@ -0,0 +1,9 @@
+import os
+import tensorflow as tf
+
+N_BOXES = 8
+default_log_dir = os.path.join(os.getcwd(), "log")
+config = tf.ConfigProto()
+config.gpu_options.allow_growth = False
+session = tf.Session(config = config)
+
diff --git a/demo/custom_vgg19.py b/demo/custom_vgg19.py
new file mode 100644
index 00000000..ec94a994
--- /dev/null
+++ b/demo/custom_vgg19.py
@@ -0,0 +1,229 @@
+# Copyright 2015 The TensorFlow Authors. All Rights Reserved.
+#
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+# ==============================================================================
+# pylint: disable=invalid-name
+"""VGG19 model for Keras.
+
+# Reference
+
+- [Very Deep Convolutional Networks for Large-Scale Image
+Recognition](https://arxiv.org/abs/1409.1556)
+
+"""
+from __future__ import absolute_import
+from __future__ import division
+from __future__ import print_function
+
+import warnings
+
+from tensorflow.contrib.keras.python.keras import backend as K
+from tensorflow.contrib.keras.python.keras.applications.imagenet_utils import _obtain_input_shape
+from tensorflow.contrib.keras.python.keras.applications.imagenet_utils import decode_predictions  # pylint: disable=unused-import
+from tensorflow.contrib.keras.python.keras.applications.imagenet_utils import preprocess_input  # pylint: disable=unused-import
+from tensorflow.contrib.keras.python.keras.engine.topology import get_source_inputs
+from tensorflow.contrib.keras.python.keras.layers import Conv2D
+from tensorflow.contrib.keras.python.keras.layers import Dense
+from tensorflow.contrib.keras.python.keras.layers import Flatten
+from tensorflow.contrib.keras.python.keras.layers import GlobalAveragePooling2D
+from tensorflow.contrib.keras.python.keras.layers import GlobalMaxPooling2D
+from tensorflow.contrib.keras.python.keras.layers import Input
+from tensorflow.contrib.keras.python.keras.layers import AveragePooling2D as MaxPooling2D
+from tensorflow.contrib.keras.python.keras.models import Model
+from tensorflow.contrib.keras.python.keras.utils import layer_utils
+from tensorflow.contrib.keras.python.keras.utils.data_utils import get_file
+
+
+WEIGHTS_PATH = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg19_weights_tf_dim_ordering_tf_kernels.h5'
+WEIGHTS_PATH_NO_TOP = 'https://github.com/fchollet/deep-learning-models/releases/download/v0.1/vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5'
+
+
+def VGG19(include_top=True,
+          weights='imagenet',
+          input_tensor=None,
+          input_shape=None,
+          pooling=None,
+          classes=1000):
+  """Instantiates the VGG19 architecture.
+
+  Optionally loads weights pre-trained
+  on ImageNet. Note that when using TensorFlow,
+  for best performance you should set
+  `image_data_format="channels_last"` in your Keras config
+  at ~/.keras/keras.json.
+
+  The model and the weights are compatible with both
+  TensorFlow and Theano. The data format
+  convention used by the model is the one
+  specified in your Keras config file.
+
+  Arguments:
+      include_top: whether to include the 3 fully-connected
+          layers at the top of the network.
+      weights: one of `None` (random initialization)
+          or "imagenet" (pre-training on ImageNet).
+      input_tensor: optional Keras tensor (i.e. output of `layers.Input()`)
+          to use as image input for the model.
+      input_shape: optional shape tuple, only to be specified
+          if `include_top` is False (otherwise the input shape
+          has to be `(224, 224, 3)` (with `channels_last` data format)
+          or `(3, 224, 244)` (with `channels_first` data format).
+          It should have exactly 3 inputs channels,
+          and width and height should be no smaller than 48.
+          E.g. `(200, 200, 3)` would be one valid value.
+      pooling: Optional pooling mode for feature extraction
+          when `include_top` is `False`.
+          - `None` means that the output of the model will be
+              the 4D tensor output of the
+              last convolutional layer.
+          - `avg` means that global average pooling
+              will be applied to the output of the
+              last convolutional layer, and thus
+              the output of the model will be a 2D tensor.
+          - `max` means that global max pooling will
+              be applied.
+      classes: optional number of classes to classify images
+          into, only to be specified if `include_top` is True, and
+          if no `weights` argument is specified.
+
+  Returns:
+      A Keras model instance.
+
+  Raises:
+      ValueError: in case of invalid argument for `weights`,
+          or invalid input shape.
+  """
+  if weights not in {'imagenet', None}:
+    raise ValueError('The `weights` argument should be either '
+                     '`None` (random initialization) or `imagenet` '
+                     '(pre-training on ImageNet).')
+
+  if weights == 'imagenet' and include_top and classes != 1000:
+    raise ValueError('If using `weights` as imagenet with `include_top`'
+                     ' as true, `classes` should be 1000')
+  # Determine proper input shape
+  input_shape = _obtain_input_shape(
+      input_shape,
+      default_size=224,
+      min_size=48,
+      data_format=K.image_data_format(),
+      include_top=include_top)
+
+  if input_tensor is None:
+    img_input = Input(shape=input_shape)
+  else:
+    img_input = Input(tensor=input_tensor, shape=input_shape)
+
+  # Block 1
+  x = Conv2D(
+      64, (3, 3), activation='relu', padding='same',
+      name='block1_conv1')(img_input)
+  x = Conv2D(
+      64, (3, 3), activation='relu', padding='same', name='block1_conv2')(x)
+  x = MaxPooling2D((2, 2), strides=(2, 2), name='block1_pool')(x)
+
+  # Block 2
+  x = Conv2D(
+      128, (3, 3), activation='relu', padding='same', name='block2_conv1')(x)
+  x = Conv2D(
+      128, (3, 3), activation='relu', padding='same', name='block2_conv2')(x)
+  x = MaxPooling2D((2, 2), strides=(2, 2), name='block2_pool')(x)
+
+  # Block 3
+  x = Conv2D(
+      256, (3, 3), activation='relu', padding='same', name='block3_conv1')(x)
+  x = Conv2D(
+      256, (3, 3), activation='relu', padding='same', name='block3_conv2')(x)
+  x = Conv2D(
+      256, (3, 3), activation='relu', padding='same', name='block3_conv3')(x)
+  x = Conv2D(
+      256, (3, 3), activation='relu', padding='same', name='block3_conv4')(x)
+  x = MaxPooling2D((2, 2), strides=(2, 2), name='block3_pool')(x)
+
+  # Block 4
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block4_conv1')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block4_conv2')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block4_conv3')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block4_conv4')(x)
+  x = MaxPooling2D((2, 2), strides=(2, 2), name='block4_pool')(x)
+
+  # Block 5
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block5_conv1')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block5_conv2')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block5_conv3')(x)
+  x = Conv2D(
+      512, (3, 3), activation='relu', padding='same', name='block5_conv4')(x)
+  x = MaxPooling2D((2, 2), strides=(2, 2), name='block5_pool')(x)
+
+  if include_top:
+    # Classification block
+    x = Flatten(name='flatten')(x)
+    x = Dense(4096, activation='relu', name='fc1')(x)
+    x = Dense(4096, activation='relu', name='fc2')(x)
+    x = Dense(classes, activation='softmax', name='predictions')(x)
+  else:
+    if pooling == 'avg':
+      x = GlobalAveragePooling2D()(x)
+    elif pooling == 'max':
+      x = GlobalMaxPooling2D()(x)
+
+  # Ensure that the model takes into account
+  # any potential predecessors of `input_tensor`.
+  if input_tensor is not None:
+    inputs = get_source_inputs(input_tensor)
+  else:
+    inputs = img_input
+  # Create model.
+  model = Model(inputs, x, name='vgg19')
+
+  # load weights
+  if weights == 'imagenet':
+    if include_top:
+      weights_path = get_file(
+          'vgg19_weights_tf_dim_ordering_tf_kernels.h5',
+          WEIGHTS_PATH,
+          cache_subdir='models')
+    else:
+      weights_path = get_file(
+          'vgg19_weights_tf_dim_ordering_tf_kernels_notop.h5',
+          WEIGHTS_PATH_NO_TOP,
+          cache_subdir='models')
+    model.load_weights(weights_path)
+    if K.backend() == 'theano':
+      layer_utils.convert_all_kernels_in_model(model)
+
+    if K.image_data_format() == 'channels_first':
+      if include_top:
+        maxpool = model.get_layer(name='block5_pool')
+        shape = maxpool.output_shape[1:]
+        dense = model.get_layer(name='fc1')
+        layer_utils.convert_dense_weights_data_format(dense, shape,
+                                                      'channels_first')
+
+      if K.backend() == 'tensorflow':
+        warnings.warn('You are using the TensorFlow backend, yet you '
+                      'are using the Theano '
+                      'image data format convention '
+                      '(`image_data_format="channels_first"`). '
+                      'For best performance, set '
+                      '`image_data_format="channels_last"` in '
+                      'your Keras config '
+                      'at ~/.keras/keras.json.')
+  return model
diff --git a/demo/deeploss.py b/demo/deeploss.py
new file mode 100644
index 00000000..e18d282b
--- /dev/null
+++ b/demo/deeploss.py
@@ -0,0 +1,258 @@
+import tensorflow as tf
+import numpy as np
+# vgg19 from keras
+from tensorflow.contrib.keras.api.keras.models import Model
+#from tensorflow.contrib.keras.api.keras.applications.vgg19 import VGG19
+from custom_vgg19 import VGG19
+from tensorflow.contrib.keras.api.keras import backend as K
+import models
+
+
+def preprocess_input(x):
+    """Preprocesses a tensor encoding a batch of images.
+    # Arguments
+        x: input tensor, 4D in [-1,1]
+    # Returns
+        Preprocessed tensor.
+    """
+    # from [-1, 1] to [0,255.0]
+    x = (x + 1.0) / 2.0 * 255.0
+    # 'RGB'->'BGR'
+    x = x[:, :, :, ::-1]
+    # Zero-center by mean pixel
+    x = x - np.array([103.939, 116.779, 123.68]).reshape((1,1,1,3))
+    return x
+
+
+class VGG19Features(object):
+    def __init__(self, session, feature_layers = None, feature_weights = None):
+        K.set_session(session)
+        self.base_model = VGG19(
+                include_top = False,
+                weights='imagenet')
+        if feature_layers is None:
+            feature_layers = [
+                    "input_1",
+                    "block1_conv2", "block2_conv2",
+                    "block3_conv2", "block4_conv2",
+                    "block5_conv2"]
+        self.layer_names = [l.name for l in self.base_model.layers]
+        for k in feature_layers:
+            if not k in self.layer_names:
+                raise KeyError(
+                        "Invalid layer {}. Available layers: {}".format(
+                            k, self.layer_names))
+        features = [self.base_model.get_layer(k).output for k in feature_layers]
+        self.model = Model(
+                inputs = self.base_model.input,
+                outputs = features)
+        if feature_weights is None:
+            feature_weights = len(feature_layers) * [1.0]
+            gram_weights = len(feature_layers) * [0.1]
+        self.feature_weights = feature_weights
+        self.gram_weights = gram_weights
+        assert len(self.feature_weights) == len(features)
+
+        self.variables = self.base_model.weights
+
+
+    def extract_features(self, x):
+        """x should be rgb in [-1,1]."""
+        x = preprocess_input(x)
+        features = self.model.predict(x)
+        return features
+
+
+    def make_feature_ops(self, x):
+        """x should be rgb tensor in [-1,1]."""
+        x = preprocess_input(x)
+        features = self.model(x)
+        return features
+
+
+    def grams(self, fs):
+        gs = list()
+        for f in fs:
+            bs, h, w, c = f.shape.as_list()
+            f = tf.reshape(f, [bs, h*w, c])
+            ft = tf.transpose(f, [0,2,1])
+            g = tf.matmul(ft, f)
+            g = g / (4.0*h*w)
+            gs.append(g)
+        return gs
+
+
+    def make_loss_op(self, x, y):
+        """x, y should be rgb tensors in [-1,1]."""
+        x = preprocess_input(x)
+        x_features = self.model(x)
+
+        y = preprocess_input(y)
+        y_features = self.model(y)
+
+        x_grams = self.grams(x_features)
+        y_grams = self.grams(y_features)
+
+        losses = [
+                tf.reduce_mean(tf.abs(xf - yf)) for xf, yf in zip(
+                    x_features, y_features)]
+        gram_losses = [
+                tf.reduce_mean(tf.abs(xg - yg)) for xg, yg in zip(
+                    x_grams, y_grams)]
+
+        for i in range(len(losses)):
+            losses[i] = self.feature_weights[i] * losses[i]
+            gram_losses[i] = self.gram_weights[i] * gram_losses[i]
+        loss = tf.add_n(losses) + tf.add_n(gram_losses)
+
+        self.losses = losses
+        self.gram_losses = gram_losses
+
+        return loss
+
+
+class PixelFeatures(object):
+    def __init__(self, session):
+        self.variables = []
+
+
+    def make_loss_op(self, x, y):
+        """x, y should be rgb tensors in [-1,1]."""
+        x_features = [x]
+        y_features = [y]
+
+        losses = [
+                tf.reduce_mean(
+                    tf.reduce_sum(
+                        tf.abs(xf - yf),
+                        axis = [1,2,3]))
+                for xf, yf in zip(x_features, y_features)]
+
+        self.feature_weights = 11*[1000.0/(128*128*3)]
+        for i in range(len(losses)):
+            losses[i] = self.feature_weights[i] * losses[i]
+
+        loss = tf.add_n(losses)
+
+        self.losses = losses
+
+        return loss
+
+
+class JigsawFeatures(object):
+    def __init__(self, session):
+        self.cfn = models.make_model(
+                "cfn", models.cfn_features,
+                n_scales = 5,
+                max_filters = 256)
+
+        x_init = tf.placeholder(
+                tf.float32,
+                shape = [64,128,128,3])
+        _ = self.cfn(x_init, init = True, dropout_p = 0.5)
+
+        self.variables = [v for v in tf.trainable_variables() if
+                v.name.startswith("cfn")]
+        self.saver = tf.train.Saver(self.variables)
+        restore_path = "../jigsaw/log/2017-08-16T14:27:04/checkpoints/model.ckpt-100000"
+        self.saver.restore(session, restore_path)
+
+        self.kwargs = {"init": False, "dropout_p": 0.0}
+
+
+    def make_loss_op(self, x, y):
+        """x, y should be rgb tensors in [-1,1]."""
+        x_features = [x]
+        y_features = [y]
+        x_features += self.cfn(x, **self.kwargs)
+        y_features += self.cfn(y, **self.kwargs)
+
+        losses = [
+                tf.reduce_mean(
+                    tf.reduce_sum(
+                        tf.abs(xf - yf),
+                        axis = [1,2,3]))
+                for xf, yf in zip(x_features, y_features)]
+
+        self.feature_weights = 11*[1.0/(128*128*3)]
+        for i in range(len(losses)):
+            losses[i] = self.feature_weights[i] * losses[i]
+
+        loss = tf.add_n(losses)
+
+        self.losses = losses
+
+        return loss
+
+
+class AttrFeatures(object):
+    def __init__(self, session):
+        self.fnet = models.make_model(
+                "encoder", models.feature_encoder,
+                n_scales = 5,
+                max_filters = 512)
+
+        x_init = tf.placeholder(
+                tf.float32,
+                shape = [64,128,128,3])
+        _ = self.fnet(x_init, init = True, dropout_p = 0.5)
+
+        self.variables = [v for v in tf.trainable_variables() if
+                v.name.startswith("encoder")]
+        self.saver = tf.train.Saver(self.variables)
+        restore_path = "log/2017-08-13T21:33:12/checkpoints/model.ckpt-100000"
+        self.saver.restore(session, restore_path)
+
+        self.kwargs = {"init": False, "dropout_p": 0.0}
+
+
+    def make_loss_op(self, x, y):
+        """x, y should be rgb tensors in [-1,1]."""
+        x_features = [x]
+        y_features = [y]
+        x_features += self.fnet(x, **self.kwargs)
+        y_features += self.fnet(y, **self.kwargs)
+
+        losses = [
+                tf.reduce_mean(
+                    tf.reduce_sum(
+                        tf.abs(xf - yf),
+                        axis = [1,2,3]))
+                for xf, yf in zip(x_features, y_features)]
+
+        self.feature_weights = 11*[1.0/(128*128*3)]
+        for i in range(len(losses)):
+            losses[i] = self.feature_weights[i] * losses[i]
+
+        loss = tf.add_n(losses)
+
+        self.losses = losses
+
+        return loss
+
+
+if __name__ == "__main__":
+    import sys
+    from tensorflow.contrib.keras.api.keras.preprocessing import (
+            image)
+
+    s = tf.Session()
+
+    img_path = sys.argv[1]
+    img = image.load_img(img_path, target_size=(128, 128))
+    x = image.img_to_array(img)
+    x = np.expand_dims(x, axis=0)
+    x = x / 255.0 * 2.0 - 1.0
+    print(x.shape, np.min(x), np.max(x))
+    x = tf.constant(x)
+
+    feature_layers = [
+            "input_1", "block1_conv1", "block1_conv2", "block1_pool", "block2_conv2",
+            "block3_conv2", "block4_conv2", "block5_conv2"]
+    vgg19 = VGG19Features(feature_layers)
+    fmaps = vgg19.make_feature_ops(x)
+
+    for i in range(len(fmaps)):
+        print(i)
+        f = fmaps[i].eval(session=s)
+        print(f.shape)
diff --git a/demo/events.out.tfevents.1509984274.hcigpu02 b/demo/events.out.tfevents.1509984274.hcigpu02
new file mode 100644
index 00000000..23a3425a
Binary files /dev/null and b/demo/events.out.tfevents.1509984274.hcigpu02 differ
diff --git a/demo/get_batches.py b/demo/get_batches.py
new file mode 100644
index 00000000..c94a2526
--- /dev/null
+++ b/demo/get_batches.py
@@ -0,0 +1,52 @@
+
+import sys
+from batches_pg2 import plot_batch
+from index_flow import IndexFlow
+from buffered_wrapper import BufferedWrapper
+
+def get_batches(
+        shape,
+        index_path,
+        train,
+        mask,
+        fill_batches = True,
+        shuffle = True,
+        return_keys = ["imgs", "joints", "norm_imgs", "norm_joints"]):
+    """Buffered IndexFlow."""
+    flow = IndexFlow(shape, index_path, train, mask, fill_batches, shuffle, return_keys)
+    return BufferedWrapper(flow)
+
+
+if __name__ == "__main__":
+    if not len(sys.argv) == 2:
+        print("Useage: {} <path to index.p>".format(sys.argv[0]))
+        exit(1)
+
+    batches = get_batches(
+            shape = (16, 128, 128, 3),
+            index_path = sys.argv[1],
+            train = True,
+            mask = False,
+            shuffle = True)
+    X, C = next(batches)
+    plot_batch(X, "unmasked.png")
+    plot_batch(C, "joints.png")
+
+    """
+    batches = get_batches(
+            shape = (16, 128, 128, 3),
+            index_path = sys.argv[1],
+            train = True,
+            mask = True)
+    X, C = next(batches)
+    plot_batch(X, "masked.png")
+
+    batches = get_batches(
+            shape = (16, 32, 32, 3),
+            index_path = sys.argv[1],
+            train = True,
+            mask = True)
+    X, C = next(batches)
+    plot_batch(X, "masked32.png")
+    plot_batch(C, "joints32.png")
+    """
diff --git a/demo/index_flow.py b/demo/index_flow.py
new file mode 100644
index 00000000..ad480490
--- /dev/null
+++ b/demo/index_flow.py
@@ -0,0 +1,138 @@
+import numpy as np
+import pickle
+import os
+
+from batches_pg2 import (
+    valid_joints,
+    load_img,
+    preprocess,
+    preprocess_mask,
+    make_joint_img,
+    normalize,
+    make_mask_img
+)
+
+class IndexFlow(object):
+    """Batches from index file."""
+
+    def __init__(
+            self,
+            shape,
+            index_path,
+            train,
+            mask = True,
+            fill_batches = True,
+            shuffle = True,
+            return_keys = ["imgs", "joints"]):
+        self.shape = shape
+        self.batch_size = self.shape[0]
+        self.img_shape = self.shape[1:]
+        with open(index_path, "rb") as f:
+            self.index = pickle.load(f)
+        self.basepath = os.path.dirname(index_path)
+        self.train = train
+        self.mask = mask
+        self.fill_batches = fill_batches
+        self.shuffle_ = shuffle
+        self.return_keys = return_keys
+
+        self.jo = self.index["joint_order"]
+        # rescale joint coordinates to image shape
+        h,w = self.img_shape[:2]
+        wh = np.array([[[w,h]]])
+        self.index["joints"] = self.index["joints"] * wh
+
+        self.indices = np.array(
+                [i for i in range(len(self.index["train"]))
+                    if self._filter(i)])
+
+        self.n = self.indices.shape[0]
+        self.shuffle()
+
+
+    def _filter(self, i):
+        good = True
+        good = good and (self.index["train"][i] == self.train)
+        joints = self.index["joints"][i]
+        required_joints = ["lshoulder","rshoulder","lhip","rhip"]
+        joint_indices = [self.jo.index(b) for b in required_joints]
+        joints = np.float32(joints[joint_indices])
+        good = good and valid_joints(joints)
+        return good
+
+
+    def __next__(self):
+        batch = dict()
+
+        # get indices for batch
+        batch_start, batch_end = self.batch_start, self.batch_start + self.batch_size
+        batch_indices = self.indices[batch_start:batch_end]
+        if self.fill_batches and batch_indices.shape[0] != self.batch_size:
+            n_missing = self.batch_size - batch_indices.shape[0]
+            batch_indices = np.concatenate([batch_indices, self.indices[:n_missing]], axis = 0)
+            assert(batch_indices.shape[0] == self.batch_size)
+        batch_indices = np.array(batch_indices)
+        batch["indices"] = batch_indices
+
+        # prepare next batch
+        if batch_end >= self.n:
+            self.shuffle()
+        else:
+            self.batch_start = batch_end
+
+        # prepare batch data
+        # load images
+        batch["imgs"] = list()
+        for i in batch_indices:
+            fname = self.index["imgs"][i]
+            # traintest = "train" if self.train else "test"
+            # path = os.path.join(self.basepath, "..", "original", "filted_up_{}".format(traintest), fname)
+            path = os.path.join(self.basepath, fname)
+            batch["imgs"].append(load_img(path, target_size = self.img_shape))
+        batch["imgs"] = np.stack(batch["imgs"])
+        batch["imgs"] = preprocess(batch["imgs"])
+
+        # load joint coordinates
+        batch["joints_coordinates"] = [self.index["joints"][i] for i in batch_indices]
+
+        # generate stickmen images from coordinates
+        batch["joints"] = list()
+        for joints in batch["joints_coordinates"]:
+            img = make_joint_img(self.img_shape, self.jo, joints)
+            batch["joints"].append(img)
+        batch["joints"] = np.stack(batch["joints"])
+        batch["joints"] = preprocess(batch["joints"])
+
+        if False and self.mask:
+            if "masks" in self.index:
+                batch_masks = list()
+                for i in batch_indices:
+                    fname = self.index["masks"][i]
+                    path = os.path.join(self.basepath, fname)
+                    batch_masks.append(load_img(path, target_size = self.img_shape))
+            else:
+                # generate mask based on joint coordinates
+                batch_masks = list()
+                for joints in batch["joints_coordinates"]:
+                    mask = make_mask_img(self.img_shape, self.jo, joints)
+                    batch_masks.append(mask)
+            batch["masks"] = np.stack(batch_masks)
+            batch["masks"] = preprocess_mask(batch["masks"])
+            # apply mask to images
+            batch["imgs"] = batch["imgs"] * batch["masks"]
+
+
+        imgs, joints = normalize(batch["imgs"], batch["joints_coordinates"], batch["joints"], self.jo)
+        batch["norm_imgs"] = imgs
+        batch["norm_joints"] = joints
+
+        batch_list = [batch[k] for k in self.return_keys]
+        return batch_list
+
+
+    def shuffle(self):
+        self.batch_start = 0
+        if self.shuffle_:
+            np.random.shuffle(self.indices)
+
+
diff --git a/demo/log.txt b/demo/log.txt
new file mode 100644
index 00000000..97ec97c4
--- /dev/null
+++ b/demo/log.txt
@@ -0,0 +1,7307 @@
+INFO:__main__:Namespace(batch_size=8, checkpoint='log/2017-10-25T16:31:50/checkpoints/model.ckpt-96000', ckpt_freq=1000, data_index='/export/home/pesser/groupstorage/data/PG2/processed00/index.p', drop_prob=0.1, init_batches=4, log_dir='/net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log', log_freq=250, lr=0.001, lr_decay_begin=1000, lr_decay_end=100000, mask=True, mode='train', spatial_size=256, test_freq=1000)
+INFO:__main__:Number of training samples: 29365
+INFO:__main__:Number of validation samples: 7336
+INFO:__main__:Latent shape: [8, 4, 4, 128]
+INFO:__main__:Latent shape: [8, 8, 8, 128]
+INFO:__main__:Defined graph
+INFO:__main__:Restored model from log/2017-10-25T16:31:50/checkpoints/model.ckpt-96000
+INFO:__main__:global_step: 1
+INFO:__main__:kl_loss: 60.20652770996094
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2023.456787109375
+INFO:__main__:loss: 2023.456787109375
+INFO:__main__:vgg_gram_loss_0: 12.149802207946777
+INFO:__main__:vgg_gram_loss_1: 61.064796447753906
+INFO:__main__:vgg_gram_loss_2: 59.43148422241211
+INFO:__main__:vgg_gram_loss_3: 35.10826873779297
+INFO:__main__:vgg_gram_loss_4: 100.04723358154297
+INFO:__main__:vgg_gram_loss_5: 0.1606256514787674
+INFO:__main__:vgg_loss_0: 10.296666145324707
+INFO:__main__:vgg_loss_1: 22.996049880981445
+INFO:__main__:vgg_loss_2: 32.64899444580078
+INFO:__main__:vgg_loss_3: 24.495223999023438
+INFO:__main__:vgg_loss_4: 44.76327133178711
+INFO:__main__:vgg_loss_5: 1.5289274454116821
+INFO:__main__:validation_loss: 3355.6640625
+INFO:__main__:global_step: 251
+INFO:__main__:kl_loss: 173.14788818359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1666.423583984375
+INFO:__main__:loss: 1666.4237060546875
+INFO:__main__:vgg_gram_loss_0: 20.077381134033203
+INFO:__main__:vgg_gram_loss_1: 32.40482711791992
+INFO:__main__:vgg_gram_loss_2: 27.869312286376953
+INFO:__main__:vgg_gram_loss_3: 20.718557357788086
+INFO:__main__:vgg_gram_loss_4: 78.97044372558594
+INFO:__main__:vgg_gram_loss_5: 0.16276900470256805
+INFO:__main__:vgg_loss_0: 13.030125617980957
+INFO:__main__:vgg_loss_1: 25.78565788269043
+INFO:__main__:vgg_loss_2: 35.54685974121094
+INFO:__main__:vgg_loss_3: 27.092498779296875
+INFO:__main__:vgg_loss_4: 49.93857955932617
+INFO:__main__:vgg_loss_5: 1.687692642211914
+INFO:__main__:validation_loss: 1993.6649169921875
+INFO:__main__:global_step: 501
+INFO:__main__:kl_loss: 265.14068603515625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2187.02294921875
+INFO:__main__:loss: 2187.023193359375
+INFO:__main__:vgg_gram_loss_0: 10.400809288024902
+INFO:__main__:vgg_gram_loss_1: 39.22432327270508
+INFO:__main__:vgg_gram_loss_2: 39.60877227783203
+INFO:__main__:vgg_gram_loss_3: 32.36015319824219
+INFO:__main__:vgg_gram_loss_4: 115.88433837890625
+INFO:__main__:vgg_gram_loss_5: 0.19620363414287567
+INFO:__main__:vgg_loss_0: 15.687895774841309
+INFO:__main__:vgg_loss_1: 35.55717849731445
+INFO:__main__:vgg_loss_2: 48.08674240112305
+INFO:__main__:vgg_loss_3: 35.310001373291016
+INFO:__main__:vgg_loss_4: 62.99497604370117
+INFO:__main__:vgg_loss_5: 2.0931894779205322
+INFO:__main__:validation_loss: 2363.730224609375
+INFO:__main__:global_step: 751
+INFO:__main__:kl_loss: 340.3248291015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1956.13134765625
+INFO:__main__:loss: 1956.1317138671875
+INFO:__main__:vgg_gram_loss_0: 18.76248550415039
+INFO:__main__:vgg_gram_loss_1: 41.23588180541992
+INFO:__main__:vgg_gram_loss_2: 38.775596618652344
+INFO:__main__:vgg_gram_loss_3: 26.62955665588379
+INFO:__main__:vgg_gram_loss_4: 94.28778076171875
+INFO:__main__:vgg_gram_loss_5: 0.16194427013397217
+INFO:__main__:vgg_loss_0: 14.774397850036621
+INFO:__main__:vgg_loss_1: 30.749303817749023
+INFO:__main__:vgg_loss_2: 41.24767303466797
+INFO:__main__:vgg_loss_3: 30.020872116088867
+INFO:__main__:vgg_loss_4: 52.83884811401367
+INFO:__main__:vgg_loss_5: 1.7419308423995972
+INFO:__main__:validation_loss: 2045.3822021484375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 1001
+INFO:__main__:kl_loss: 364.91424560546875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1847.637939453125
+INFO:__main__:loss: 1847.6383056640625
+INFO:__main__:vgg_gram_loss_0: 13.954872131347656
+INFO:__main__:vgg_gram_loss_1: 35.536136627197266
+INFO:__main__:vgg_gram_loss_2: 31.152179718017578
+INFO:__main__:vgg_gram_loss_3: 25.605749130249023
+INFO:__main__:vgg_gram_loss_4: 96.90245056152344
+INFO:__main__:vgg_gram_loss_5: 0.18925856053829193
+INFO:__main__:vgg_loss_0: 13.134907722473145
+INFO:__main__:vgg_loss_1: 30.432283401489258
+INFO:__main__:vgg_loss_2: 39.65081024169922
+INFO:__main__:vgg_loss_3: 28.924427032470703
+INFO:__main__:vgg_loss_4: 52.24839401245117
+INFO:__main__:vgg_loss_5: 1.7961026430130005
+INFO:__main__:validation_loss: 2097.765869140625
+INFO:__main__:global_step: 1251
+INFO:__main__:kl_loss: 398.1080627441406
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1941.419677734375
+INFO:__main__:loss: 1941.4200439453125
+INFO:__main__:vgg_gram_loss_0: 18.83348274230957
+INFO:__main__:vgg_gram_loss_1: 33.11129379272461
+INFO:__main__:vgg_gram_loss_2: 29.547454833984375
+INFO:__main__:vgg_gram_loss_3: 22.545629501342773
+INFO:__main__:vgg_gram_loss_4: 94.59038543701172
+INFO:__main__:vgg_gram_loss_5: 0.17973926663398743
+INFO:__main__:vgg_loss_0: 15.769280433654785
+INFO:__main__:vgg_loss_1: 32.75959777832031
+INFO:__main__:vgg_loss_2: 45.70271301269531
+INFO:__main__:vgg_loss_3: 33.16744613647461
+INFO:__main__:vgg_loss_4: 60.09803009033203
+INFO:__main__:vgg_loss_5: 1.9788700342178345
+INFO:__main__:validation_loss: 2015.1007080078125
+INFO:__main__:global_step: 1501
+INFO:__main__:kl_loss: 516.0904541015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1786.6876220703125
+INFO:__main__:loss: 1786.6881103515625
+INFO:__main__:vgg_gram_loss_0: 14.205286026000977
+INFO:__main__:vgg_gram_loss_1: 27.378259658813477
+INFO:__main__:vgg_gram_loss_2: 33.25638198852539
+INFO:__main__:vgg_gram_loss_3: 23.698928833007812
+INFO:__main__:vgg_gram_loss_4: 87.22620391845703
+INFO:__main__:vgg_gram_loss_5: 0.16732317209243774
+INFO:__main__:vgg_loss_0: 13.700509071350098
+INFO:__main__:vgg_loss_1: 30.18804359436035
+INFO:__main__:vgg_loss_2: 41.626312255859375
+INFO:__main__:vgg_loss_3: 30.353256225585938
+INFO:__main__:vgg_loss_4: 53.790164947509766
+INFO:__main__:vgg_loss_5: 1.746876835823059
+INFO:__main__:validation_loss: 2140.793701171875
+INFO:__main__:global_step: 1751
+INFO:__main__:kl_loss: 517.904052734375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1886.168212890625
+INFO:__main__:loss: 1886.168701171875
+INFO:__main__:vgg_gram_loss_0: 18.17301368713379
+INFO:__main__:vgg_gram_loss_1: 44.91527557373047
+INFO:__main__:vgg_gram_loss_2: 36.753273010253906
+INFO:__main__:vgg_gram_loss_3: 25.398351669311523
+INFO:__main__:vgg_gram_loss_4: 94.37198638916016
+INFO:__main__:vgg_gram_loss_5: 0.16551266610622406
+INFO:__main__:vgg_loss_0: 11.549256324768066
+INFO:__main__:vgg_loss_1: 25.97317886352539
+INFO:__main__:vgg_loss_2: 37.60198974609375
+INFO:__main__:vgg_loss_3: 28.477455139160156
+INFO:__main__:vgg_loss_4: 52.14436340332031
+INFO:__main__:vgg_loss_5: 1.7099798917770386
+INFO:__main__:validation_loss: 2123.537841796875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 2001
+INFO:__main__:kl_loss: 724.5712280273438
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2327.536865234375
+INFO:__main__:loss: 2327.53759765625
+INFO:__main__:vgg_gram_loss_0: 17.133886337280273
+INFO:__main__:vgg_gram_loss_1: 59.6475715637207
+INFO:__main__:vgg_gram_loss_2: 41.20440673828125
+INFO:__main__:vgg_gram_loss_3: 34.825626373291016
+INFO:__main__:vgg_gram_loss_4: 111.4234848022461
+INFO:__main__:vgg_gram_loss_5: 0.21071158349514008
+INFO:__main__:vgg_loss_0: 16.122833251953125
+INFO:__main__:vgg_loss_1: 39.45900344848633
+INFO:__main__:vgg_loss_2: 48.76203918457031
+INFO:__main__:vgg_loss_3: 34.72369384765625
+INFO:__main__:vgg_loss_4: 59.98519515991211
+INFO:__main__:vgg_loss_5: 2.008913993835449
+INFO:__main__:validation_loss: 2361.58251953125
+INFO:__main__:global_step: 2251
+INFO:__main__:kl_loss: 701.630126953125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2170.956298828125
+INFO:__main__:loss: 2170.95703125
+INFO:__main__:vgg_gram_loss_0: 17.107728958129883
+INFO:__main__:vgg_gram_loss_1: 36.949127197265625
+INFO:__main__:vgg_gram_loss_2: 36.3306770324707
+INFO:__main__:vgg_gram_loss_3: 31.79855728149414
+INFO:__main__:vgg_gram_loss_4: 114.5477523803711
+INFO:__main__:vgg_gram_loss_5: 0.2067674845457077
+INFO:__main__:vgg_loss_0: 16.155357360839844
+INFO:__main__:vgg_loss_1: 33.5197868347168
+INFO:__main__:vgg_loss_2: 47.23255157470703
+INFO:__main__:vgg_loss_3: 35.08713150024414
+INFO:__main__:vgg_loss_4: 63.1865119934082
+INFO:__main__:vgg_loss_5: 2.0693235397338867
+INFO:__main__:validation_loss: 1757.99365234375
+INFO:__main__:global_step: 2501
+INFO:__main__:kl_loss: 813.126953125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1948.814697265625
+INFO:__main__:loss: 1948.8155517578125
+INFO:__main__:vgg_gram_loss_0: 13.936678886413574
+INFO:__main__:vgg_gram_loss_1: 32.551326751708984
+INFO:__main__:vgg_gram_loss_2: 30.84714698791504
+INFO:__main__:vgg_gram_loss_3: 24.257200241088867
+INFO:__main__:vgg_gram_loss_4: 98.40998077392578
+INFO:__main__:vgg_gram_loss_5: 0.19481153786182404
+INFO:__main__:vgg_loss_0: 14.516462326049805
+INFO:__main__:vgg_loss_1: 31.28249740600586
+INFO:__main__:vgg_loss_2: 45.52101516723633
+INFO:__main__:vgg_loss_3: 34.17013168334961
+INFO:__main__:vgg_loss_4: 61.984283447265625
+INFO:__main__:vgg_loss_5: 2.0914158821105957
+INFO:__main__:validation_loss: 2176.75439453125
+INFO:__main__:global_step: 2751
+INFO:__main__:kl_loss: 876.168212890625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1803.515380859375
+INFO:__main__:loss: 1803.5162353515625
+INFO:__main__:vgg_gram_loss_0: 19.57283592224121
+INFO:__main__:vgg_gram_loss_1: 41.463653564453125
+INFO:__main__:vgg_gram_loss_2: 34.83030319213867
+INFO:__main__:vgg_gram_loss_3: 23.803367614746094
+INFO:__main__:vgg_gram_loss_4: 88.15823364257812
+INFO:__main__:vgg_gram_loss_5: 0.16864289343357086
+INFO:__main__:vgg_loss_0: 12.400886535644531
+INFO:__main__:vgg_loss_1: 26.177406311035156
+INFO:__main__:vgg_loss_2: 36.32575225830078
+INFO:__main__:vgg_loss_3: 27.017194747924805
+INFO:__main__:vgg_loss_4: 49.116207122802734
+INFO:__main__:vgg_loss_5: 1.6685727834701538
+INFO:__main__:validation_loss: 2109.174560546875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 3001
+INFO:__main__:kl_loss: 874.681396484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1993.727783203125
+INFO:__main__:loss: 1993.7286376953125
+INFO:__main__:vgg_gram_loss_0: 9.39432430267334
+INFO:__main__:vgg_gram_loss_1: 49.747406005859375
+INFO:__main__:vgg_gram_loss_2: 40.818172454833984
+INFO:__main__:vgg_gram_loss_3: 28.133386611938477
+INFO:__main__:vgg_gram_loss_4: 92.25126647949219
+INFO:__main__:vgg_gram_loss_5: 0.1733321249485016
+INFO:__main__:vgg_loss_0: 14.199066162109375
+INFO:__main__:vgg_loss_1: 33.83330535888672
+INFO:__main__:vgg_loss_2: 42.886558532714844
+INFO:__main__:vgg_loss_3: 30.94998550415039
+INFO:__main__:vgg_loss_4: 54.553794860839844
+INFO:__main__:vgg_loss_5: 1.8049606084823608
+INFO:__main__:validation_loss: 1760.3115234375
+INFO:__main__:global_step: 3251
+INFO:__main__:kl_loss: 832.4716186523438
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1796.2969970703125
+INFO:__main__:loss: 1796.2978515625
+INFO:__main__:vgg_gram_loss_0: 13.52221393585205
+INFO:__main__:vgg_gram_loss_1: 29.603153228759766
+INFO:__main__:vgg_gram_loss_2: 26.941171646118164
+INFO:__main__:vgg_gram_loss_3: 22.704923629760742
+INFO:__main__:vgg_gram_loss_4: 89.89064025878906
+INFO:__main__:vgg_gram_loss_5: 0.1799435317516327
+INFO:__main__:vgg_loss_0: 13.937674522399902
+INFO:__main__:vgg_loss_1: 31.861469268798828
+INFO:__main__:vgg_loss_2: 42.74617004394531
+INFO:__main__:vgg_loss_3: 30.874834060668945
+INFO:__main__:vgg_loss_4: 55.147701263427734
+INFO:__main__:vgg_loss_5: 1.849489450454712
+INFO:__main__:validation_loss: 2409.61962890625
+INFO:__main__:global_step: 3501
+INFO:__main__:kl_loss: 928.4254150390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1850.1971435546875
+INFO:__main__:loss: 1850.1981201171875
+INFO:__main__:vgg_gram_loss_0: 18.71107292175293
+INFO:__main__:vgg_gram_loss_1: 30.63387107849121
+INFO:__main__:vgg_gram_loss_2: 28.906375885009766
+INFO:__main__:vgg_gram_loss_3: 23.24925422668457
+INFO:__main__:vgg_gram_loss_4: 93.26155853271484
+INFO:__main__:vgg_gram_loss_5: 0.19740276038646698
+INFO:__main__:vgg_loss_0: 13.973001480102539
+INFO:__main__:vgg_loss_1: 29.986095428466797
+INFO:__main__:vgg_loss_2: 42.69750213623047
+INFO:__main__:vgg_loss_3: 31.014432907104492
+INFO:__main__:vgg_loss_4: 55.51621627807617
+INFO:__main__:vgg_loss_5: 1.8926173448562622
+INFO:__main__:validation_loss: 1825.427734375
+INFO:__main__:global_step: 3751
+INFO:__main__:kl_loss: 1118.3740234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1970.6109619140625
+INFO:__main__:loss: 1970.612060546875
+INFO:__main__:vgg_gram_loss_0: 13.173460960388184
+INFO:__main__:vgg_gram_loss_1: 39.9528694152832
+INFO:__main__:vgg_gram_loss_2: 34.67822265625
+INFO:__main__:vgg_gram_loss_3: 26.1641845703125
+INFO:__main__:vgg_gram_loss_4: 106.43123626708984
+INFO:__main__:vgg_gram_loss_5: 0.18200846016407013
+INFO:__main__:vgg_loss_0: 13.137272834777832
+INFO:__main__:vgg_loss_1: 31.23973274230957
+INFO:__main__:vgg_loss_2: 41.655094146728516
+INFO:__main__:vgg_loss_3: 30.921598434448242
+INFO:__main__:vgg_loss_4: 54.81023406982422
+INFO:__main__:vgg_loss_5: 1.7762824296951294
+INFO:__main__:validation_loss: 2355.878662109375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 4001
+INFO:__main__:kl_loss: 1141.31201171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1659.1766357421875
+INFO:__main__:loss: 1659.177734375
+INFO:__main__:vgg_gram_loss_0: 15.296905517578125
+INFO:__main__:vgg_gram_loss_1: 29.74534034729004
+INFO:__main__:vgg_gram_loss_2: 26.242666244506836
+INFO:__main__:vgg_gram_loss_3: 19.549055099487305
+INFO:__main__:vgg_gram_loss_4: 84.08551788330078
+INFO:__main__:vgg_gram_loss_5: 0.1774490922689438
+INFO:__main__:vgg_loss_0: 11.836042404174805
+INFO:__main__:vgg_loss_1: 25.887290954589844
+INFO:__main__:vgg_loss_2: 37.43425369262695
+INFO:__main__:vgg_loss_3: 28.02289581298828
+INFO:__main__:vgg_loss_4: 51.81277847290039
+INFO:__main__:vgg_loss_5: 1.7451210021972656
+INFO:__main__:validation_loss: 2517.59521484375
+INFO:__main__:global_step: 4251
+INFO:__main__:kl_loss: 1199.059814453125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2222.4951171875
+INFO:__main__:loss: 2222.496337890625
+INFO:__main__:vgg_gram_loss_0: 14.938468933105469
+INFO:__main__:vgg_gram_loss_1: 40.962032318115234
+INFO:__main__:vgg_gram_loss_2: 33.206573486328125
+INFO:__main__:vgg_gram_loss_3: 36.72642135620117
+INFO:__main__:vgg_gram_loss_4: 111.7970962524414
+INFO:__main__:vgg_gram_loss_5: 0.2122902125120163
+INFO:__main__:vgg_loss_0: 17.67517852783203
+INFO:__main__:vgg_loss_1: 39.86125564575195
+INFO:__main__:vgg_loss_2: 47.52143096923828
+INFO:__main__:vgg_loss_3: 35.75613784790039
+INFO:__main__:vgg_loss_4: 63.707515716552734
+INFO:__main__:vgg_loss_5: 2.134636163711548
+INFO:__main__:validation_loss: 1828.6475830078125
+INFO:__main__:global_step: 4501
+INFO:__main__:kl_loss: 1247.668212890625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1640.203369140625
+INFO:__main__:loss: 1640.20458984375
+INFO:__main__:vgg_gram_loss_0: 12.333768844604492
+INFO:__main__:vgg_gram_loss_1: 20.17340660095215
+INFO:__main__:vgg_gram_loss_2: 23.368614196777344
+INFO:__main__:vgg_gram_loss_3: 19.387720108032227
+INFO:__main__:vgg_gram_loss_4: 90.00624084472656
+INFO:__main__:vgg_gram_loss_5: 0.1868402510881424
+INFO:__main__:vgg_loss_0: 11.576725959777832
+INFO:__main__:vgg_loss_1: 24.910717010498047
+INFO:__main__:vgg_loss_2: 38.58959197998047
+INFO:__main__:vgg_loss_3: 29.907052993774414
+INFO:__main__:vgg_loss_4: 55.688236236572266
+INFO:__main__:vgg_loss_5: 1.9117759466171265
+INFO:__main__:validation_loss: 1857.666748046875
+INFO:__main__:global_step: 4751
+INFO:__main__:kl_loss: 1330.21435546875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1772.7276611328125
+INFO:__main__:loss: 1772.72900390625
+INFO:__main__:vgg_gram_loss_0: 12.628029823303223
+INFO:__main__:vgg_gram_loss_1: 30.191160202026367
+INFO:__main__:vgg_gram_loss_2: 28.742584228515625
+INFO:__main__:vgg_gram_loss_3: 23.408655166625977
+INFO:__main__:vgg_gram_loss_4: 94.96575164794922
+INFO:__main__:vgg_gram_loss_5: 0.17650341987609863
+INFO:__main__:vgg_loss_0: 12.080496788024902
+INFO:__main__:vgg_loss_1: 26.644556045532227
+INFO:__main__:vgg_loss_2: 39.07718276977539
+INFO:__main__:vgg_loss_3: 29.733915328979492
+INFO:__main__:vgg_loss_4: 55.0627555847168
+INFO:__main__:vgg_loss_5: 1.8339576721191406
+INFO:__main__:validation_loss: 1741.0556640625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 5001
+INFO:__main__:kl_loss: 1371.990234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2060.505859375
+INFO:__main__:loss: 2060.50732421875
+INFO:__main__:vgg_gram_loss_0: 10.114445686340332
+INFO:__main__:vgg_gram_loss_1: 27.754196166992188
+INFO:__main__:vgg_gram_loss_2: 42.70623779296875
+INFO:__main__:vgg_gram_loss_3: 31.155282974243164
+INFO:__main__:vgg_gram_loss_4: 106.2396469116211
+INFO:__main__:vgg_gram_loss_5: 0.2120528221130371
+INFO:__main__:vgg_loss_0: 13.734189987182617
+INFO:__main__:vgg_loss_1: 33.53791809082031
+INFO:__main__:vgg_loss_2: 47.36857604980469
+INFO:__main__:vgg_loss_3: 34.70136642456055
+INFO:__main__:vgg_loss_4: 62.465728759765625
+INFO:__main__:vgg_loss_5: 2.1115729808807373
+INFO:__main__:validation_loss: 2520.490966796875
+INFO:__main__:global_step: 5251
+INFO:__main__:kl_loss: 1243.628662109375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1891.0390625
+INFO:__main__:loss: 1891.040283203125
+INFO:__main__:vgg_gram_loss_0: 12.999078750610352
+INFO:__main__:vgg_gram_loss_1: 43.352413177490234
+INFO:__main__:vgg_gram_loss_2: 32.902950286865234
+INFO:__main__:vgg_gram_loss_3: 23.427854537963867
+INFO:__main__:vgg_gram_loss_4: 87.47883605957031
+INFO:__main__:vgg_gram_loss_5: 0.18693867325782776
+INFO:__main__:vgg_loss_0: 13.056669235229492
+INFO:__main__:vgg_loss_1: 30.94521713256836
+INFO:__main__:vgg_loss_2: 43.70237731933594
+INFO:__main__:vgg_loss_3: 31.98383331298828
+INFO:__main__:vgg_loss_4: 56.30106735229492
+INFO:__main__:vgg_loss_5: 1.8705779314041138
+INFO:__main__:validation_loss: 1730.5810546875
+INFO:__main__:global_step: 5501
+INFO:__main__:kl_loss: 1728.80224609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2488.67724609375
+INFO:__main__:loss: 2488.678955078125
+INFO:__main__:vgg_gram_loss_0: 21.705032348632812
+INFO:__main__:vgg_gram_loss_1: 71.9174575805664
+INFO:__main__:vgg_gram_loss_2: 54.15242385864258
+INFO:__main__:vgg_gram_loss_3: 38.18333053588867
+INFO:__main__:vgg_gram_loss_4: 113.03267669677734
+INFO:__main__:vgg_gram_loss_5: 0.21321962773799896
+INFO:__main__:vgg_loss_0: 16.03223991394043
+INFO:__main__:vgg_loss_1: 37.315208435058594
+INFO:__main__:vgg_loss_2: 48.6584358215332
+INFO:__main__:vgg_loss_3: 35.010501861572266
+INFO:__main__:vgg_loss_4: 59.5318489074707
+INFO:__main__:vgg_loss_5: 1.98308265209198
+INFO:__main__:validation_loss: 2196.510498046875
+INFO:__main__:global_step: 5751
+INFO:__main__:kl_loss: 1414.7266845703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2082.0341796875
+INFO:__main__:loss: 2082.03564453125
+INFO:__main__:vgg_gram_loss_0: 17.04506492614746
+INFO:__main__:vgg_gram_loss_1: 40.56757736206055
+INFO:__main__:vgg_gram_loss_2: 39.39512252807617
+INFO:__main__:vgg_gram_loss_3: 28.267248153686523
+INFO:__main__:vgg_gram_loss_4: 101.11149597167969
+INFO:__main__:vgg_gram_loss_5: 0.1915002018213272
+INFO:__main__:vgg_loss_0: 15.066765785217285
+INFO:__main__:vgg_loss_1: 32.68598937988281
+INFO:__main__:vgg_loss_2: 45.99399948120117
+INFO:__main__:vgg_loss_3: 33.79623794555664
+INFO:__main__:vgg_loss_4: 60.278602600097656
+INFO:__main__:vgg_loss_5: 2.007242441177368
+INFO:__main__:validation_loss: 2262.83984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 6001
+INFO:__main__:kl_loss: 1736.0665283203125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2166.789794921875
+INFO:__main__:loss: 2166.79150390625
+INFO:__main__:vgg_gram_loss_0: 12.11079216003418
+INFO:__main__:vgg_gram_loss_1: 51.705543518066406
+INFO:__main__:vgg_gram_loss_2: 32.836151123046875
+INFO:__main__:vgg_gram_loss_3: 32.321495056152344
+INFO:__main__:vgg_gram_loss_4: 110.64031219482422
+INFO:__main__:vgg_gram_loss_5: 0.19285158812999725
+INFO:__main__:vgg_loss_0: 15.51778507232666
+INFO:__main__:vgg_loss_1: 35.28180694580078
+INFO:__main__:vgg_loss_2: 45.66849899291992
+INFO:__main__:vgg_loss_3: 34.265281677246094
+INFO:__main__:vgg_loss_4: 60.84768295288086
+INFO:__main__:vgg_loss_5: 1.9697527885437012
+INFO:__main__:validation_loss: 2238.836181640625
+INFO:__main__:global_step: 6251
+INFO:__main__:kl_loss: 1911.279296875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2473.839111328125
+INFO:__main__:loss: 2473.841064453125
+INFO:__main__:vgg_gram_loss_0: 10.262408256530762
+INFO:__main__:vgg_gram_loss_1: 44.55028533935547
+INFO:__main__:vgg_gram_loss_2: 53.020362854003906
+INFO:__main__:vgg_gram_loss_3: 38.87364959716797
+INFO:__main__:vgg_gram_loss_4: 124.38823699951172
+INFO:__main__:vgg_gram_loss_5: 0.255666583776474
+INFO:__main__:vgg_loss_0: 17.760690689086914
+INFO:__main__:vgg_loss_1: 43.78571701049805
+INFO:__main__:vgg_loss_2: 55.12312698364258
+INFO:__main__:vgg_loss_3: 38.84754943847656
+INFO:__main__:vgg_loss_4: 65.6886215209961
+INFO:__main__:vgg_loss_5: 2.211482524871826
+INFO:__main__:validation_loss: 1960.843994140625
+INFO:__main__:global_step: 6501
+INFO:__main__:kl_loss: 1598.666748046875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1777.1851806640625
+INFO:__main__:loss: 1777.186767578125
+INFO:__main__:vgg_gram_loss_0: 15.061113357543945
+INFO:__main__:vgg_gram_loss_1: 34.10643005371094
+INFO:__main__:vgg_gram_loss_2: 29.44182777404785
+INFO:__main__:vgg_gram_loss_3: 21.689485549926758
+INFO:__main__:vgg_gram_loss_4: 85.99014282226562
+INFO:__main__:vgg_gram_loss_5: 0.17214810848236084
+INFO:__main__:vgg_loss_0: 13.499688148498535
+INFO:__main__:vgg_loss_1: 29.6932373046875
+INFO:__main__:vgg_loss_2: 40.63449478149414
+INFO:__main__:vgg_loss_3: 29.72477912902832
+INFO:__main__:vgg_loss_4: 53.668426513671875
+INFO:__main__:vgg_loss_5: 1.7552709579467773
+INFO:__main__:validation_loss: 1826.8505859375
+INFO:__main__:global_step: 6751
+INFO:__main__:kl_loss: 2119.90771484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1971.783447265625
+INFO:__main__:loss: 1971.7855224609375
+INFO:__main__:vgg_gram_loss_0: 16.287397384643555
+INFO:__main__:vgg_gram_loss_1: 27.925281524658203
+INFO:__main__:vgg_gram_loss_2: 33.74029541015625
+INFO:__main__:vgg_gram_loss_3: 26.581327438354492
+INFO:__main__:vgg_gram_loss_4: 97.57390594482422
+INFO:__main__:vgg_gram_loss_5: 0.1936752200126648
+INFO:__main__:vgg_loss_0: 14.744640350341797
+INFO:__main__:vgg_loss_1: 32.6653938293457
+INFO:__main__:vgg_loss_2: 47.16722106933594
+INFO:__main__:vgg_loss_3: 34.90412521362305
+INFO:__main__:vgg_loss_4: 60.58540725708008
+INFO:__main__:vgg_loss_5: 1.9879920482635498
+INFO:__main__:validation_loss: 2444.9775390625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 7001
+INFO:__main__:kl_loss: 1759.365478515625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2016.00927734375
+INFO:__main__:loss: 2016.010986328125
+INFO:__main__:vgg_gram_loss_0: 20.884267807006836
+INFO:__main__:vgg_gram_loss_1: 40.27299880981445
+INFO:__main__:vgg_gram_loss_2: 42.23915481567383
+INFO:__main__:vgg_gram_loss_3: 30.190610885620117
+INFO:__main__:vgg_gram_loss_4: 94.78038787841797
+INFO:__main__:vgg_gram_loss_5: 0.19803155958652496
+INFO:__main__:vgg_loss_0: 13.698533058166504
+INFO:__main__:vgg_loss_1: 31.094709396362305
+INFO:__main__:vgg_loss_2: 41.62931823730469
+INFO:__main__:vgg_loss_3: 30.673473358154297
+INFO:__main__:vgg_loss_4: 55.59429168701172
+INFO:__main__:vgg_loss_5: 1.9460351467132568
+INFO:__main__:validation_loss: 1596.9501953125
+INFO:__main__:global_step: 7251
+INFO:__main__:kl_loss: 1937.2467041015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1885.02197265625
+INFO:__main__:loss: 1885.02392578125
+INFO:__main__:vgg_gram_loss_0: 18.52371597290039
+INFO:__main__:vgg_gram_loss_1: 37.670501708984375
+INFO:__main__:vgg_gram_loss_2: 32.39860153198242
+INFO:__main__:vgg_gram_loss_3: 23.78447151184082
+INFO:__main__:vgg_gram_loss_4: 87.36138916015625
+INFO:__main__:vgg_gram_loss_5: 0.18844832479953766
+INFO:__main__:vgg_loss_0: 14.654539108276367
+INFO:__main__:vgg_loss_1: 31.725788116455078
+INFO:__main__:vgg_loss_2: 42.622432708740234
+INFO:__main__:vgg_loss_3: 31.314496994018555
+INFO:__main__:vgg_loss_4: 54.91584014892578
+INFO:__main__:vgg_loss_5: 1.8441699743270874
+INFO:__main__:validation_loss: 1773.9703369140625
+INFO:__main__:global_step: 7501
+INFO:__main__:kl_loss: 2131.0380859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2021.41015625
+INFO:__main__:loss: 2021.4122314453125
+INFO:__main__:vgg_gram_loss_0: 13.283345222473145
+INFO:__main__:vgg_gram_loss_1: 43.86565017700195
+INFO:__main__:vgg_gram_loss_2: 32.67824935913086
+INFO:__main__:vgg_gram_loss_3: 24.61185646057129
+INFO:__main__:vgg_gram_loss_4: 96.58019256591797
+INFO:__main__:vgg_gram_loss_5: 0.19690720736980438
+INFO:__main__:vgg_loss_0: 15.022104263305664
+INFO:__main__:vgg_loss_1: 34.14169692993164
+INFO:__main__:vgg_loss_2: 46.85837173461914
+INFO:__main__:vgg_loss_3: 34.1616325378418
+INFO:__main__:vgg_loss_4: 60.87163162231445
+INFO:__main__:vgg_loss_5: 2.0103940963745117
+INFO:__main__:validation_loss: 2252.0986328125
+INFO:__main__:global_step: 7751
+INFO:__main__:kl_loss: 2289.4501953125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1883.4287109375
+INFO:__main__:loss: 1883.4310302734375
+INFO:__main__:vgg_gram_loss_0: 6.607985019683838
+INFO:__main__:vgg_gram_loss_1: 36.47885513305664
+INFO:__main__:vgg_gram_loss_2: 29.464889526367188
+INFO:__main__:vgg_gram_loss_3: 21.269906997680664
+INFO:__main__:vgg_gram_loss_4: 89.96973419189453
+INFO:__main__:vgg_gram_loss_5: 0.18583011627197266
+INFO:__main__:vgg_loss_0: 15.688158988952637
+INFO:__main__:vgg_loss_1: 37.999820709228516
+INFO:__main__:vgg_loss_2: 47.02214431762695
+INFO:__main__:vgg_loss_3: 33.02042770385742
+INFO:__main__:vgg_loss_4: 57.17068862915039
+INFO:__main__:vgg_loss_5: 1.8072850704193115
+INFO:__main__:validation_loss: 1997.0162353515625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 8001
+INFO:__main__:kl_loss: 2434.55322265625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2146.66064453125
+INFO:__main__:loss: 2146.6630859375
+INFO:__main__:vgg_gram_loss_0: 7.746669292449951
+INFO:__main__:vgg_gram_loss_1: 43.344051361083984
+INFO:__main__:vgg_gram_loss_2: 32.98564147949219
+INFO:__main__:vgg_gram_loss_3: 31.324813842773438
+INFO:__main__:vgg_gram_loss_4: 108.05046844482422
+INFO:__main__:vgg_gram_loss_5: 0.2084517925977707
+INFO:__main__:vgg_loss_0: 16.95317840576172
+INFO:__main__:vgg_loss_1: 38.46986770629883
+INFO:__main__:vgg_loss_2: 47.86206817626953
+INFO:__main__:vgg_loss_3: 36.16668701171875
+INFO:__main__:vgg_loss_4: 64.12487030029297
+INFO:__main__:vgg_loss_5: 2.095412015914917
+INFO:__main__:validation_loss: 1841.358154296875
+INFO:__main__:global_step: 8251
+INFO:__main__:kl_loss: 2044.19677734375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1917.5123291015625
+INFO:__main__:loss: 1917.514404296875
+INFO:__main__:vgg_gram_loss_0: 15.649087905883789
+INFO:__main__:vgg_gram_loss_1: 30.912633895874023
+INFO:__main__:vgg_gram_loss_2: 30.14988136291504
+INFO:__main__:vgg_gram_loss_3: 27.8271541595459
+INFO:__main__:vgg_gram_loss_4: 98.56961059570312
+INFO:__main__:vgg_gram_loss_5: 0.1898769736289978
+INFO:__main__:vgg_loss_0: 14.73673152923584
+INFO:__main__:vgg_loss_1: 32.670494079589844
+INFO:__main__:vgg_loss_2: 43.859127044677734
+INFO:__main__:vgg_loss_3: 31.57451057434082
+INFO:__main__:vgg_loss_4: 55.51279067993164
+INFO:__main__:vgg_loss_5: 1.8505703210830688
+INFO:__main__:validation_loss: 1790.004150390625
+INFO:__main__:global_step: 8501
+INFO:__main__:kl_loss: 2297.97119140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1941.687255859375
+INFO:__main__:loss: 1941.6895751953125
+INFO:__main__:vgg_gram_loss_0: 11.816027641296387
+INFO:__main__:vgg_gram_loss_1: 28.37053871154785
+INFO:__main__:vgg_gram_loss_2: 34.72297668457031
+INFO:__main__:vgg_gram_loss_3: 25.13543701171875
+INFO:__main__:vgg_gram_loss_4: 95.15686798095703
+INFO:__main__:vgg_gram_loss_5: 0.18899141252040863
+INFO:__main__:vgg_loss_0: 14.185519218444824
+INFO:__main__:vgg_loss_1: 33.80742263793945
+INFO:__main__:vgg_loss_2: 48.83255386352539
+INFO:__main__:vgg_loss_3: 34.67539596557617
+INFO:__main__:vgg_loss_4: 59.532249450683594
+INFO:__main__:vgg_loss_5: 1.9135000705718994
+INFO:__main__:validation_loss: 2503.92333984375
+INFO:__main__:global_step: 8751
+INFO:__main__:kl_loss: 2029.747802734375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1929.013427734375
+INFO:__main__:loss: 1929.0155029296875
+INFO:__main__:vgg_gram_loss_0: 13.485989570617676
+INFO:__main__:vgg_gram_loss_1: 32.256019592285156
+INFO:__main__:vgg_gram_loss_2: 31.45692253112793
+INFO:__main__:vgg_gram_loss_3: 26.671667098999023
+INFO:__main__:vgg_gram_loss_4: 97.23084259033203
+INFO:__main__:vgg_gram_loss_5: 0.17051078379154205
+INFO:__main__:vgg_loss_0: 13.79652214050293
+INFO:__main__:vgg_loss_1: 31.372127532958984
+INFO:__main__:vgg_loss_2: 44.68638229370117
+INFO:__main__:vgg_loss_3: 33.30086135864258
+INFO:__main__:vgg_loss_4: 59.520938873291016
+INFO:__main__:vgg_loss_5: 1.8539001941680908
+INFO:__main__:validation_loss: 1828.245849609375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 9001
+INFO:__main__:kl_loss: 2077.443359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1662.433837890625
+INFO:__main__:loss: 1662.4359130859375
+INFO:__main__:vgg_gram_loss_0: 20.073820114135742
+INFO:__main__:vgg_gram_loss_1: 25.09209442138672
+INFO:__main__:vgg_gram_loss_2: 23.551029205322266
+INFO:__main__:vgg_gram_loss_3: 19.17599105834961
+INFO:__main__:vgg_gram_loss_4: 81.62151336669922
+INFO:__main__:vgg_gram_loss_5: 0.1748962700366974
+INFO:__main__:vgg_loss_0: 12.520403861999512
+INFO:__main__:vgg_loss_1: 27.71816062927246
+INFO:__main__:vgg_loss_2: 39.48741912841797
+INFO:__main__:vgg_loss_3: 29.41246223449707
+INFO:__main__:vgg_loss_4: 51.911136627197266
+INFO:__main__:vgg_loss_5: 1.747796893119812
+INFO:__main__:validation_loss: 2209.35205078125
+INFO:__main__:global_step: 9251
+INFO:__main__:kl_loss: 2461.293212890625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1896.51416015625
+INFO:__main__:loss: 1896.5166015625
+INFO:__main__:vgg_gram_loss_0: 8.450826644897461
+INFO:__main__:vgg_gram_loss_1: 32.96268081665039
+INFO:__main__:vgg_gram_loss_2: 35.25413131713867
+INFO:__main__:vgg_gram_loss_3: 28.76320457458496
+INFO:__main__:vgg_gram_loss_4: 105.91483306884766
+INFO:__main__:vgg_gram_loss_5: 0.1938469111919403
+INFO:__main__:vgg_loss_0: 12.521793365478516
+INFO:__main__:vgg_loss_1: 30.456708908081055
+INFO:__main__:vgg_loss_2: 40.558189392089844
+INFO:__main__:vgg_loss_3: 29.686697006225586
+INFO:__main__:vgg_loss_4: 52.71685791015625
+INFO:__main__:vgg_loss_5: 1.8230597972869873
+INFO:__main__:validation_loss: 1830.0399169921875
+INFO:__main__:global_step: 9501
+INFO:__main__:kl_loss: 2555.31396484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1965.73828125
+INFO:__main__:loss: 1965.7408447265625
+INFO:__main__:vgg_gram_loss_0: 13.696782112121582
+INFO:__main__:vgg_gram_loss_1: 48.06203842163086
+INFO:__main__:vgg_gram_loss_2: 36.55930709838867
+INFO:__main__:vgg_gram_loss_3: 28.607547760009766
+INFO:__main__:vgg_gram_loss_4: 95.61166381835938
+INFO:__main__:vgg_gram_loss_5: 0.165425643324852
+INFO:__main__:vgg_loss_0: 14.141860961914062
+INFO:__main__:vgg_loss_1: 31.857946395874023
+INFO:__main__:vgg_loss_2: 41.4349479675293
+INFO:__main__:vgg_loss_3: 29.938541412353516
+INFO:__main__:vgg_loss_4: 51.42768096923828
+INFO:__main__:vgg_loss_5: 1.6439051628112793
+INFO:__main__:validation_loss: 1679.08935546875
+INFO:__main__:global_step: 9751
+INFO:__main__:kl_loss: 2549.347900390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1886.56103515625
+INFO:__main__:loss: 1886.5635986328125
+INFO:__main__:vgg_gram_loss_0: 11.207842826843262
+INFO:__main__:vgg_gram_loss_1: 33.38605880737305
+INFO:__main__:vgg_gram_loss_2: 29.325515747070312
+INFO:__main__:vgg_gram_loss_3: 22.65707778930664
+INFO:__main__:vgg_gram_loss_4: 96.7847900390625
+INFO:__main__:vgg_gram_loss_5: 0.22450366616249084
+INFO:__main__:vgg_loss_0: 13.013879776000977
+INFO:__main__:vgg_loss_1: 31.942190170288086
+INFO:__main__:vgg_loss_2: 45.62629699707031
+INFO:__main__:vgg_loss_3: 33.09357833862305
+INFO:__main__:vgg_loss_4: 58.058319091796875
+INFO:__main__:vgg_loss_5: 1.9921537637710571
+INFO:__main__:validation_loss: 1996.9764404296875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 10001
+INFO:__main__:kl_loss: 3241.4482421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1702.4114990234375
+INFO:__main__:loss: 1702.414794921875
+INFO:__main__:vgg_gram_loss_0: 11.898480415344238
+INFO:__main__:vgg_gram_loss_1: 24.88400650024414
+INFO:__main__:vgg_gram_loss_2: 27.841232299804688
+INFO:__main__:vgg_gram_loss_3: 21.28725814819336
+INFO:__main__:vgg_gram_loss_4: 86.39045715332031
+INFO:__main__:vgg_gram_loss_5: 0.18358160555362701
+INFO:__main__:vgg_loss_0: 12.072579383850098
+INFO:__main__:vgg_loss_1: 27.72849464416504
+INFO:__main__:vgg_loss_2: 42.09772872924805
+INFO:__main__:vgg_loss_3: 30.993335723876953
+INFO:__main__:vgg_loss_4: 53.33146667480469
+INFO:__main__:vgg_loss_5: 1.7736963033676147
+INFO:__main__:validation_loss: 2375.9248046875
+INFO:__main__:global_step: 10251
+INFO:__main__:kl_loss: 2913.326904296875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2261.784423828125
+INFO:__main__:loss: 2261.787353515625
+INFO:__main__:vgg_gram_loss_0: 20.29518699645996
+INFO:__main__:vgg_gram_loss_1: 46.13532638549805
+INFO:__main__:vgg_gram_loss_2: 39.92474365234375
+INFO:__main__:vgg_gram_loss_3: 27.63934898376465
+INFO:__main__:vgg_gram_loss_4: 108.5341567993164
+INFO:__main__:vgg_gram_loss_5: 0.2337137907743454
+INFO:__main__:vgg_loss_0: 16.388893127441406
+INFO:__main__:vgg_loss_1: 36.86563491821289
+INFO:__main__:vgg_loss_2: 50.898094177246094
+INFO:__main__:vgg_loss_3: 37.15837860107422
+INFO:__main__:vgg_loss_4: 66.07582092285156
+INFO:__main__:vgg_loss_5: 2.2075746059417725
+INFO:__main__:validation_loss: 2008.4005126953125
+INFO:__main__:global_step: 10501
+INFO:__main__:kl_loss: 2772.4619140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1665.134033203125
+INFO:__main__:loss: 1665.1368408203125
+INFO:__main__:vgg_gram_loss_0: 17.528867721557617
+INFO:__main__:vgg_gram_loss_1: 26.647199630737305
+INFO:__main__:vgg_gram_loss_2: 26.195438385009766
+INFO:__main__:vgg_gram_loss_3: 19.85382080078125
+INFO:__main__:vgg_gram_loss_4: 85.17772674560547
+INFO:__main__:vgg_gram_loss_5: 0.17175838351249695
+INFO:__main__:vgg_loss_0: 12.163930892944336
+INFO:__main__:vgg_loss_1: 28.144418716430664
+INFO:__main__:vgg_loss_2: 39.30435562133789
+INFO:__main__:vgg_loss_3: 27.597774505615234
+INFO:__main__:vgg_loss_4: 48.574119567871094
+INFO:__main__:vgg_loss_5: 1.6673803329467773
+INFO:__main__:validation_loss: 2206.001708984375
+INFO:__main__:global_step: 10751
+INFO:__main__:kl_loss: 2703.9130859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1694.342529296875
+INFO:__main__:loss: 1694.34521484375
+INFO:__main__:vgg_gram_loss_0: 13.905160903930664
+INFO:__main__:vgg_gram_loss_1: 29.344024658203125
+INFO:__main__:vgg_gram_loss_2: 28.814090728759766
+INFO:__main__:vgg_gram_loss_3: 20.8415470123291
+INFO:__main__:vgg_gram_loss_4: 81.15296173095703
+INFO:__main__:vgg_gram_loss_5: 0.16132643818855286
+INFO:__main__:vgg_loss_0: 12.874180793762207
+INFO:__main__:vgg_loss_1: 29.179365158081055
+INFO:__main__:vgg_loss_2: 39.6439094543457
+INFO:__main__:vgg_loss_3: 28.951364517211914
+INFO:__main__:vgg_loss_4: 52.2606086730957
+INFO:__main__:vgg_loss_5: 1.7399569749832153
+INFO:__main__:validation_loss: 2034.6087646484375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 11001
+INFO:__main__:kl_loss: 2643.982177734375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1947.2125244140625
+INFO:__main__:loss: 1947.2152099609375
+INFO:__main__:vgg_gram_loss_0: 14.827585220336914
+INFO:__main__:vgg_gram_loss_1: 26.005048751831055
+INFO:__main__:vgg_gram_loss_2: 25.515039443969727
+INFO:__main__:vgg_gram_loss_3: 22.021961212158203
+INFO:__main__:vgg_gram_loss_4: 97.12137603759766
+INFO:__main__:vgg_gram_loss_5: 0.1919902265071869
+INFO:__main__:vgg_loss_0: 16.557048797607422
+INFO:__main__:vgg_loss_1: 34.50715637207031
+INFO:__main__:vgg_loss_2: 47.035057067871094
+INFO:__main__:vgg_loss_3: 36.387908935546875
+INFO:__main__:vgg_loss_4: 67.1322250366211
+INFO:__main__:vgg_loss_5: 2.1401286125183105
+INFO:__main__:validation_loss: 2133.8974609375
+INFO:__main__:global_step: 11251
+INFO:__main__:kl_loss: 3058.58642578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1920.107177734375
+INFO:__main__:loss: 1920.1102294921875
+INFO:__main__:vgg_gram_loss_0: 12.681962013244629
+INFO:__main__:vgg_gram_loss_1: 27.328372955322266
+INFO:__main__:vgg_gram_loss_2: 31.3234920501709
+INFO:__main__:vgg_gram_loss_3: 25.278867721557617
+INFO:__main__:vgg_gram_loss_4: 100.98329162597656
+INFO:__main__:vgg_gram_loss_5: 0.20000572502613068
+INFO:__main__:vgg_loss_0: 14.080020904541016
+INFO:__main__:vgg_loss_1: 32.73818588256836
+INFO:__main__:vgg_loss_2: 45.41073989868164
+INFO:__main__:vgg_loss_3: 33.08762741088867
+INFO:__main__:vgg_loss_4: 58.95177459716797
+INFO:__main__:vgg_loss_5: 1.9571093320846558
+INFO:__main__:validation_loss: 2126.898193359375
+INFO:__main__:global_step: 11501
+INFO:__main__:kl_loss: 3223.32080078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1886.1070556640625
+INFO:__main__:loss: 1886.1102294921875
+INFO:__main__:vgg_gram_loss_0: 6.771231174468994
+INFO:__main__:vgg_gram_loss_1: 31.556737899780273
+INFO:__main__:vgg_gram_loss_2: 27.337650299072266
+INFO:__main__:vgg_gram_loss_3: 21.93885612487793
+INFO:__main__:vgg_gram_loss_4: 96.59512329101562
+INFO:__main__:vgg_gram_loss_5: 0.20657996833324432
+INFO:__main__:vgg_loss_0: 14.741159439086914
+INFO:__main__:vgg_loss_1: 32.30353546142578
+INFO:__main__:vgg_loss_2: 46.44126510620117
+INFO:__main__:vgg_loss_3: 34.73818588256836
+INFO:__main__:vgg_loss_4: 62.52718734741211
+INFO:__main__:vgg_loss_5: 2.0638880729675293
+INFO:__main__:validation_loss: 2119.157470703125
+INFO:__main__:global_step: 11751
+INFO:__main__:kl_loss: 2798.33642578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1423.464599609375
+INFO:__main__:loss: 1423.4674072265625
+INFO:__main__:vgg_gram_loss_0: 9.188904762268066
+INFO:__main__:vgg_gram_loss_1: 17.212066650390625
+INFO:__main__:vgg_gram_loss_2: 19.176837921142578
+INFO:__main__:vgg_gram_loss_3: 15.346638679504395
+INFO:__main__:vgg_gram_loss_4: 70.6548080444336
+INFO:__main__:vgg_gram_loss_5: 0.16639554500579834
+INFO:__main__:vgg_loss_0: 11.862112045288086
+INFO:__main__:vgg_loss_1: 26.231782913208008
+INFO:__main__:vgg_loss_2: 37.56650161743164
+INFO:__main__:vgg_loss_3: 27.383419036865234
+INFO:__main__:vgg_loss_4: 48.24058151245117
+INFO:__main__:vgg_loss_5: 1.662850260734558
+INFO:__main__:validation_loss: 1706.93994140625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 12001
+INFO:__main__:kl_loss: 2654.115234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1618.133056640625
+INFO:__main__:loss: 1618.1357421875
+INFO:__main__:vgg_gram_loss_0: 12.687799453735352
+INFO:__main__:vgg_gram_loss_1: 25.796537399291992
+INFO:__main__:vgg_gram_loss_2: 25.808584213256836
+INFO:__main__:vgg_gram_loss_3: 20.099903106689453
+INFO:__main__:vgg_gram_loss_4: 79.86431121826172
+INFO:__main__:vgg_gram_loss_5: 0.16555766761302948
+INFO:__main__:vgg_loss_0: 11.580936431884766
+INFO:__main__:vgg_loss_1: 28.09764862060547
+INFO:__main__:vgg_loss_2: 39.67683410644531
+INFO:__main__:vgg_loss_3: 28.344676971435547
+INFO:__main__:vgg_loss_4: 49.84782791137695
+INFO:__main__:vgg_loss_5: 1.6559890508651733
+INFO:__main__:validation_loss: 1994.832763671875
+INFO:__main__:global_step: 12251
+INFO:__main__:kl_loss: 2799.060302734375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1468.264404296875
+INFO:__main__:loss: 1468.2672119140625
+INFO:__main__:vgg_gram_loss_0: 19.166973114013672
+INFO:__main__:vgg_gram_loss_1: 24.330476760864258
+INFO:__main__:vgg_gram_loss_2: 20.55609893798828
+INFO:__main__:vgg_gram_loss_3: 16.42759895324707
+INFO:__main__:vgg_gram_loss_4: 68.98406219482422
+INFO:__main__:vgg_gram_loss_5: 0.15285027027130127
+INFO:__main__:vgg_loss_0: 10.890030860900879
+INFO:__main__:vgg_loss_1: 23.132923126220703
+INFO:__main__:vgg_loss_2: 34.75040817260742
+INFO:__main__:vgg_loss_3: 26.280176162719727
+INFO:__main__:vgg_loss_4: 47.39338302612305
+INFO:__main__:vgg_loss_5: 1.5879415273666382
+INFO:__main__:validation_loss: 2264.0859375
+INFO:__main__:global_step: 12501
+INFO:__main__:kl_loss: 3204.81298828125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2331.565185546875
+INFO:__main__:loss: 2331.568359375
+INFO:__main__:vgg_gram_loss_0: 10.541520118713379
+INFO:__main__:vgg_gram_loss_1: 48.631736755371094
+INFO:__main__:vgg_gram_loss_2: 48.326087951660156
+INFO:__main__:vgg_gram_loss_3: 33.58705139160156
+INFO:__main__:vgg_gram_loss_4: 115.24056243896484
+INFO:__main__:vgg_gram_loss_5: 0.22202351689338684
+INFO:__main__:vgg_loss_0: 16.212562561035156
+INFO:__main__:vgg_loss_1: 38.53394317626953
+INFO:__main__:vgg_loss_2: 51.86149215698242
+INFO:__main__:vgg_loss_3: 36.8984260559082
+INFO:__main__:vgg_loss_4: 64.13957977294922
+INFO:__main__:vgg_loss_5: 2.118056535720825
+INFO:__main__:validation_loss: 2171.927734375
+INFO:__main__:global_step: 12751
+INFO:__main__:kl_loss: 3591.6015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1858.23876953125
+INFO:__main__:loss: 1858.2423095703125
+INFO:__main__:vgg_gram_loss_0: 8.33130168914795
+INFO:__main__:vgg_gram_loss_1: 40.207374572753906
+INFO:__main__:vgg_gram_loss_2: 31.791385650634766
+INFO:__main__:vgg_gram_loss_3: 25.225576400756836
+INFO:__main__:vgg_gram_loss_4: 90.53115844726562
+INFO:__main__:vgg_gram_loss_5: 0.18880970776081085
+INFO:__main__:vgg_loss_0: 13.582549095153809
+INFO:__main__:vgg_loss_1: 32.469207763671875
+INFO:__main__:vgg_loss_2: 41.88188171386719
+INFO:__main__:vgg_loss_3: 31.225648880004883
+INFO:__main__:vgg_loss_4: 54.397743225097656
+INFO:__main__:vgg_loss_5: 1.8151251077651978
+INFO:__main__:validation_loss: 2385.468505859375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 13001
+INFO:__main__:kl_loss: 3349.11669921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1828.945068359375
+INFO:__main__:loss: 1828.9483642578125
+INFO:__main__:vgg_gram_loss_0: 13.481892585754395
+INFO:__main__:vgg_gram_loss_1: 49.339473724365234
+INFO:__main__:vgg_gram_loss_2: 30.41879653930664
+INFO:__main__:vgg_gram_loss_3: 22.178022384643555
+INFO:__main__:vgg_gram_loss_4: 80.55927276611328
+INFO:__main__:vgg_gram_loss_5: 0.14683052897453308
+INFO:__main__:vgg_loss_0: 14.550590515136719
+INFO:__main__:vgg_loss_1: 35.22194290161133
+INFO:__main__:vgg_loss_2: 40.037681579589844
+INFO:__main__:vgg_loss_3: 28.40900230407715
+INFO:__main__:vgg_loss_4: 49.83712387084961
+INFO:__main__:vgg_loss_5: 1.6083848476409912
+INFO:__main__:validation_loss: 2253.092529296875
+INFO:__main__:global_step: 13251
+INFO:__main__:kl_loss: 4117.90625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1638.808837890625
+INFO:__main__:loss: 1638.81298828125
+INFO:__main__:vgg_gram_loss_0: 16.00355339050293
+INFO:__main__:vgg_gram_loss_1: 27.079391479492188
+INFO:__main__:vgg_gram_loss_2: 25.015766143798828
+INFO:__main__:vgg_gram_loss_3: 20.76770782470703
+INFO:__main__:vgg_gram_loss_4: 82.87042236328125
+INFO:__main__:vgg_gram_loss_5: 0.16286657750606537
+INFO:__main__:vgg_loss_0: 11.67535400390625
+INFO:__main__:vgg_loss_1: 26.841936111450195
+INFO:__main__:vgg_loss_2: 36.96250534057617
+INFO:__main__:vgg_loss_3: 27.7122745513916
+INFO:__main__:vgg_loss_4: 50.94968795776367
+INFO:__main__:vgg_loss_5: 1.7202852964401245
+INFO:__main__:validation_loss: 2117.615478515625
+INFO:__main__:global_step: 13501
+INFO:__main__:kl_loss: 5110.0400390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1785.3057861328125
+INFO:__main__:loss: 1785.3109130859375
+INFO:__main__:vgg_gram_loss_0: 15.640748977661133
+INFO:__main__:vgg_gram_loss_1: 27.0262393951416
+INFO:__main__:vgg_gram_loss_2: 29.5236873626709
+INFO:__main__:vgg_gram_loss_3: 22.32098960876465
+INFO:__main__:vgg_gram_loss_4: 91.04948425292969
+INFO:__main__:vgg_gram_loss_5: 0.18062224984169006
+INFO:__main__:vgg_loss_0: 12.897849082946777
+INFO:__main__:vgg_loss_1: 29.169940948486328
+INFO:__main__:vgg_loss_2: 41.37274169921875
+INFO:__main__:vgg_loss_3: 30.79285430908203
+INFO:__main__:vgg_loss_4: 55.28400421142578
+INFO:__main__:vgg_loss_5: 1.8019770383834839
+INFO:__main__:validation_loss: 2342.46923828125
+INFO:__main__:global_step: 13751
+INFO:__main__:kl_loss: 3353.460693359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1825.2525634765625
+INFO:__main__:loss: 1825.255859375
+INFO:__main__:vgg_gram_loss_0: 15.761792182922363
+INFO:__main__:vgg_gram_loss_1: 39.96674346923828
+INFO:__main__:vgg_gram_loss_2: 30.72432518005371
+INFO:__main__:vgg_gram_loss_3: 24.176380157470703
+INFO:__main__:vgg_gram_loss_4: 85.57173919677734
+INFO:__main__:vgg_gram_loss_5: 0.1593572497367859
+INFO:__main__:vgg_loss_0: 14.906819343566895
+INFO:__main__:vgg_loss_1: 34.24696731567383
+INFO:__main__:vgg_loss_2: 40.673587799072266
+INFO:__main__:vgg_loss_3: 28.23394203186035
+INFO:__main__:vgg_loss_4: 48.96389389038086
+INFO:__main__:vgg_loss_5: 1.664943814277649
+INFO:__main__:validation_loss: 1553.1966552734375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 14001
+INFO:__main__:kl_loss: 3246.311767578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1881.316650390625
+INFO:__main__:loss: 1881.3199462890625
+INFO:__main__:vgg_gram_loss_0: 14.242349624633789
+INFO:__main__:vgg_gram_loss_1: 30.5249080657959
+INFO:__main__:vgg_gram_loss_2: 27.972055435180664
+INFO:__main__:vgg_gram_loss_3: 21.534807205200195
+INFO:__main__:vgg_gram_loss_4: 92.98632049560547
+INFO:__main__:vgg_gram_loss_5: 0.18983612954616547
+INFO:__main__:vgg_loss_0: 15.947529792785645
+INFO:__main__:vgg_loss_1: 31.471717834472656
+INFO:__main__:vgg_loss_2: 44.951847076416016
+INFO:__main__:vgg_loss_3: 33.66530990600586
+INFO:__main__:vgg_loss_4: 60.7533073425293
+INFO:__main__:vgg_loss_5: 2.0233490467071533
+INFO:__main__:validation_loss: 1748.1754150390625
+INFO:__main__:global_step: 14251
+INFO:__main__:kl_loss: 3431.954833984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1458.1470947265625
+INFO:__main__:loss: 1458.1505126953125
+INFO:__main__:vgg_gram_loss_0: 11.338159561157227
+INFO:__main__:vgg_gram_loss_1: 19.594633102416992
+INFO:__main__:vgg_gram_loss_2: 20.68074607849121
+INFO:__main__:vgg_gram_loss_3: 16.244789123535156
+INFO:__main__:vgg_gram_loss_4: 74.70944213867188
+INFO:__main__:vgg_gram_loss_5: 0.16610495746135712
+INFO:__main__:vgg_loss_0: 10.72453498840332
+INFO:__main__:vgg_loss_1: 24.973461151123047
+INFO:__main__:vgg_loss_2: 36.35271453857422
+INFO:__main__:vgg_loss_3: 26.27377700805664
+INFO:__main__:vgg_loss_4: 48.88114547729492
+INFO:__main__:vgg_loss_5: 1.689893364906311
+INFO:__main__:validation_loss: 1886.6724853515625
+INFO:__main__:global_step: 14501
+INFO:__main__:kl_loss: 3985.45361328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1910.81884765625
+INFO:__main__:loss: 1910.8228759765625
+INFO:__main__:vgg_gram_loss_0: 15.037406921386719
+INFO:__main__:vgg_gram_loss_1: 24.871139526367188
+INFO:__main__:vgg_gram_loss_2: 31.145315170288086
+INFO:__main__:vgg_gram_loss_3: 26.66287612915039
+INFO:__main__:vgg_gram_loss_4: 103.48453521728516
+INFO:__main__:vgg_gram_loss_5: 0.20381061732769012
+INFO:__main__:vgg_loss_0: 13.707566261291504
+INFO:__main__:vgg_loss_1: 29.62269401550293
+INFO:__main__:vgg_loss_2: 44.128196716308594
+INFO:__main__:vgg_loss_3: 33.2076301574707
+INFO:__main__:vgg_loss_4: 58.15578079223633
+INFO:__main__:vgg_loss_5: 1.9368208646774292
+INFO:__main__:validation_loss: 1973.97119140625
+INFO:__main__:global_step: 14751
+INFO:__main__:kl_loss: 4005.341796875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1795.0201416015625
+INFO:__main__:loss: 1795.024169921875
+INFO:__main__:vgg_gram_loss_0: 12.135330200195312
+INFO:__main__:vgg_gram_loss_1: 26.882963180541992
+INFO:__main__:vgg_gram_loss_2: 27.51361083984375
+INFO:__main__:vgg_gram_loss_3: 22.776376724243164
+INFO:__main__:vgg_gram_loss_4: 90.52733612060547
+INFO:__main__:vgg_gram_loss_5: 0.20045800507068634
+INFO:__main__:vgg_loss_0: 12.593707084655762
+INFO:__main__:vgg_loss_1: 30.002941131591797
+INFO:__main__:vgg_loss_2: 44.14188766479492
+INFO:__main__:vgg_loss_3: 31.992292404174805
+INFO:__main__:vgg_loss_4: 58.2563591003418
+INFO:__main__:vgg_loss_5: 1.980757236480713
+INFO:__main__:validation_loss: 2075.4521484375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 15001
+INFO:__main__:kl_loss: 3705.33642578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1595.1666259765625
+INFO:__main__:loss: 1595.1702880859375
+INFO:__main__:vgg_gram_loss_0: 10.975664138793945
+INFO:__main__:vgg_gram_loss_1: 22.13172721862793
+INFO:__main__:vgg_gram_loss_2: 24.078147888183594
+INFO:__main__:vgg_gram_loss_3: 19.292373657226562
+INFO:__main__:vgg_gram_loss_4: 82.76775360107422
+INFO:__main__:vgg_gram_loss_5: 0.1540568321943283
+INFO:__main__:vgg_loss_0: 11.58592700958252
+INFO:__main__:vgg_loss_1: 25.57155990600586
+INFO:__main__:vgg_loss_2: 38.18693161010742
+INFO:__main__:vgg_loss_3: 28.970317840576172
+INFO:__main__:vgg_loss_4: 53.57235336303711
+INFO:__main__:vgg_loss_5: 1.7465263605117798
+INFO:__main__:validation_loss: 1781.873046875
+INFO:__main__:global_step: 15251
+INFO:__main__:kl_loss: 3564.941650390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1503.5670166015625
+INFO:__main__:loss: 1503.570556640625
+INFO:__main__:vgg_gram_loss_0: 8.540643692016602
+INFO:__main__:vgg_gram_loss_1: 24.819456100463867
+INFO:__main__:vgg_gram_loss_2: 24.256214141845703
+INFO:__main__:vgg_gram_loss_3: 18.178815841674805
+INFO:__main__:vgg_gram_loss_4: 72.755126953125
+INFO:__main__:vgg_gram_loss_5: 0.16218386590480804
+INFO:__main__:vgg_loss_0: 11.633992195129395
+INFO:__main__:vgg_loss_1: 27.11802864074707
+INFO:__main__:vgg_loss_2: 37.86495590209961
+INFO:__main__:vgg_loss_3: 26.818740844726562
+INFO:__main__:vgg_loss_4: 46.961204528808594
+INFO:__main__:vgg_loss_5: 1.6040434837341309
+INFO:__main__:validation_loss: 1919.8582763671875
+INFO:__main__:global_step: 15501
+INFO:__main__:kl_loss: 3862.746337890625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1528.5277099609375
+INFO:__main__:loss: 1528.5316162109375
+INFO:__main__:vgg_gram_loss_0: 6.751155376434326
+INFO:__main__:vgg_gram_loss_1: 19.732168197631836
+INFO:__main__:vgg_gram_loss_2: 25.24652671813965
+INFO:__main__:vgg_gram_loss_3: 19.191333770751953
+INFO:__main__:vgg_gram_loss_4: 81.19164276123047
+INFO:__main__:vgg_gram_loss_5: 0.17378969490528107
+INFO:__main__:vgg_loss_0: 11.43883991241455
+INFO:__main__:vgg_loss_1: 27.002866744995117
+INFO:__main__:vgg_loss_2: 37.634483337402344
+INFO:__main__:vgg_loss_3: 27.2590274810791
+INFO:__main__:vgg_loss_4: 48.447044372558594
+INFO:__main__:vgg_loss_5: 1.6366742849349976
+INFO:__main__:validation_loss: 1905.0755615234375
+INFO:__main__:global_step: 15751
+INFO:__main__:kl_loss: 4190.8232421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2032.50634765625
+INFO:__main__:loss: 2032.510498046875
+INFO:__main__:vgg_gram_loss_0: 13.058337211608887
+INFO:__main__:vgg_gram_loss_1: 34.21646499633789
+INFO:__main__:vgg_gram_loss_2: 32.2791862487793
+INFO:__main__:vgg_gram_loss_3: 27.77005958557129
+INFO:__main__:vgg_gram_loss_4: 112.00274658203125
+INFO:__main__:vgg_gram_loss_5: 0.20230674743652344
+INFO:__main__:vgg_loss_0: 13.781364440917969
+INFO:__main__:vgg_loss_1: 32.421974182128906
+INFO:__main__:vgg_loss_2: 46.06599044799805
+INFO:__main__:vgg_loss_3: 33.66859817504883
+INFO:__main__:vgg_loss_4: 59.12703323364258
+INFO:__main__:vgg_loss_5: 1.9072597026824951
+INFO:__main__:validation_loss: 1991.9036865234375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 16001
+INFO:__main__:kl_loss: 3942.93701171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1856.5948486328125
+INFO:__main__:loss: 1856.5987548828125
+INFO:__main__:vgg_gram_loss_0: 14.474349975585938
+INFO:__main__:vgg_gram_loss_1: 27.639780044555664
+INFO:__main__:vgg_gram_loss_2: 26.839385986328125
+INFO:__main__:vgg_gram_loss_3: 20.698564529418945
+INFO:__main__:vgg_gram_loss_4: 91.61420440673828
+INFO:__main__:vgg_gram_loss_5: 0.1941124051809311
+INFO:__main__:vgg_loss_0: 13.904624938964844
+INFO:__main__:vgg_loss_1: 30.179750442504883
+INFO:__main__:vgg_loss_2: 45.923309326171875
+INFO:__main__:vgg_loss_3: 35.29290771484375
+INFO:__main__:vgg_loss_4: 62.52599334716797
+INFO:__main__:vgg_loss_5: 2.032017707824707
+INFO:__main__:validation_loss: 2060.87158203125
+INFO:__main__:global_step: 16251
+INFO:__main__:kl_loss: 4026.26220703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1694.01416015625
+INFO:__main__:loss: 1694.0181884765625
+INFO:__main__:vgg_gram_loss_0: 8.845500946044922
+INFO:__main__:vgg_gram_loss_1: 24.206079483032227
+INFO:__main__:vgg_gram_loss_2: 23.476261138916016
+INFO:__main__:vgg_gram_loss_3: 20.182392120361328
+INFO:__main__:vgg_gram_loss_4: 88.28746795654297
+INFO:__main__:vgg_gram_loss_5: 0.17453892529010773
+INFO:__main__:vgg_loss_0: 12.2302885055542
+INFO:__main__:vgg_loss_1: 29.776330947875977
+INFO:__main__:vgg_loss_2: 43.28388977050781
+INFO:__main__:vgg_loss_3: 31.666786193847656
+INFO:__main__:vgg_loss_4: 54.920257568359375
+INFO:__main__:vgg_loss_5: 1.753005862236023
+INFO:__main__:validation_loss: 2319.92236328125
+INFO:__main__:global_step: 16501
+INFO:__main__:kl_loss: 6287.22607421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2443.67578125
+INFO:__main__:loss: 2443.68212890625
+INFO:__main__:vgg_gram_loss_0: 10.505154609680176
+INFO:__main__:vgg_gram_loss_1: 84.68325805664062
+INFO:__main__:vgg_gram_loss_2: 35.788570404052734
+INFO:__main__:vgg_gram_loss_3: 35.972164154052734
+INFO:__main__:vgg_gram_loss_4: 109.2228775024414
+INFO:__main__:vgg_gram_loss_5: 0.2088581770658493
+INFO:__main__:vgg_loss_0: 18.007917404174805
+INFO:__main__:vgg_loss_1: 45.69658660888672
+INFO:__main__:vgg_loss_2: 48.703975677490234
+INFO:__main__:vgg_loss_3: 36.1117057800293
+INFO:__main__:vgg_loss_4: 61.785709381103516
+INFO:__main__:vgg_loss_5: 2.048408269882202
+INFO:__main__:validation_loss: 1796.233642578125
+INFO:__main__:global_step: 16751
+INFO:__main__:kl_loss: 5193.3701171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1613.5125732421875
+INFO:__main__:loss: 1613.517822265625
+INFO:__main__:vgg_gram_loss_0: 9.90428352355957
+INFO:__main__:vgg_gram_loss_1: 24.719528198242188
+INFO:__main__:vgg_gram_loss_2: 23.356409072875977
+INFO:__main__:vgg_gram_loss_3: 19.07557487487793
+INFO:__main__:vgg_gram_loss_4: 84.72393035888672
+INFO:__main__:vgg_gram_loss_5: 0.18478351831436157
+INFO:__main__:vgg_loss_0: 11.846158027648926
+INFO:__main__:vgg_loss_1: 27.13921546936035
+INFO:__main__:vgg_loss_2: 38.92344284057617
+INFO:__main__:vgg_loss_3: 28.96397590637207
+INFO:__main__:vgg_loss_4: 52.080604553222656
+INFO:__main__:vgg_loss_5: 1.7846001386642456
+INFO:__main__:validation_loss: 2220.157470703125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 17001
+INFO:__main__:kl_loss: 5167.78857421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2153.682373046875
+INFO:__main__:loss: 2153.6875
+INFO:__main__:vgg_gram_loss_0: 17.497949600219727
+INFO:__main__:vgg_gram_loss_1: 35.26213073730469
+INFO:__main__:vgg_gram_loss_2: 39.78397750854492
+INFO:__main__:vgg_gram_loss_3: 30.014745712280273
+INFO:__main__:vgg_gram_loss_4: 108.40862274169922
+INFO:__main__:vgg_gram_loss_5: 0.19994519650936127
+INFO:__main__:vgg_loss_0: 15.828639030456543
+INFO:__main__:vgg_loss_1: 35.10080337524414
+INFO:__main__:vgg_loss_2: 47.98162078857422
+INFO:__main__:vgg_loss_3: 35.97663879394531
+INFO:__main__:vgg_loss_4: 62.64408493041992
+INFO:__main__:vgg_loss_5: 2.037324905395508
+INFO:__main__:validation_loss: 1857.9671630859375
+INFO:__main__:global_step: 17251
+INFO:__main__:kl_loss: 4789.04931640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1838.816162109375
+INFO:__main__:loss: 1838.8209228515625
+INFO:__main__:vgg_gram_loss_0: 15.293060302734375
+INFO:__main__:vgg_gram_loss_1: 33.45280075073242
+INFO:__main__:vgg_gram_loss_2: 26.56186294555664
+INFO:__main__:vgg_gram_loss_3: 22.263498306274414
+INFO:__main__:vgg_gram_loss_4: 90.66622161865234
+INFO:__main__:vgg_gram_loss_5: 0.18304787576198578
+INFO:__main__:vgg_loss_0: 14.160183906555176
+INFO:__main__:vgg_loss_1: 33.23466110229492
+INFO:__main__:vgg_loss_2: 43.448856353759766
+INFO:__main__:vgg_loss_3: 31.427034378051758
+INFO:__main__:vgg_loss_4: 55.25357437133789
+INFO:__main__:vgg_loss_5: 1.8184367418289185
+INFO:__main__:validation_loss: 2144.089599609375
+INFO:__main__:global_step: 17501
+INFO:__main__:kl_loss: 4353.240234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1636.551513671875
+INFO:__main__:loss: 1636.555908203125
+INFO:__main__:vgg_gram_loss_0: 9.070740699768066
+INFO:__main__:vgg_gram_loss_1: 21.96969985961914
+INFO:__main__:vgg_gram_loss_2: 24.71034812927246
+INFO:__main__:vgg_gram_loss_3: 19.113136291503906
+INFO:__main__:vgg_gram_loss_4: 83.71758270263672
+INFO:__main__:vgg_gram_loss_5: 0.1788686066865921
+INFO:__main__:vgg_loss_0: 11.988703727722168
+INFO:__main__:vgg_loss_1: 28.854387283325195
+INFO:__main__:vgg_loss_2: 41.723995208740234
+INFO:__main__:vgg_loss_3: 30.296030044555664
+INFO:__main__:vgg_loss_4: 53.894081115722656
+INFO:__main__:vgg_loss_5: 1.7927372455596924
+INFO:__main__:validation_loss: 3001.1982421875
+INFO:__main__:global_step: 17751
+INFO:__main__:kl_loss: 4352.79736328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1734.5400390625
+INFO:__main__:loss: 1734.54443359375
+INFO:__main__:vgg_gram_loss_0: 7.867430210113525
+INFO:__main__:vgg_gram_loss_1: 27.18879508972168
+INFO:__main__:vgg_gram_loss_2: 27.73171043395996
+INFO:__main__:vgg_gram_loss_3: 22.349393844604492
+INFO:__main__:vgg_gram_loss_4: 85.84656524658203
+INFO:__main__:vgg_gram_loss_5: 0.19267073273658752
+INFO:__main__:vgg_loss_0: 13.095458030700684
+INFO:__main__:vgg_loss_1: 30.833934783935547
+INFO:__main__:vgg_loss_2: 42.64744186401367
+INFO:__main__:vgg_loss_3: 31.58111572265625
+INFO:__main__:vgg_loss_4: 55.693206787109375
+INFO:__main__:vgg_loss_5: 1.8802882432937622
+INFO:__main__:validation_loss: 1898.9388427734375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 18001
+INFO:__main__:kl_loss: 4494.47265625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1945.37939453125
+INFO:__main__:loss: 1945.3839111328125
+INFO:__main__:vgg_gram_loss_0: 13.733105659484863
+INFO:__main__:vgg_gram_loss_1: 29.994049072265625
+INFO:__main__:vgg_gram_loss_2: 28.75807762145996
+INFO:__main__:vgg_gram_loss_3: 26.727506637573242
+INFO:__main__:vgg_gram_loss_4: 104.34041595458984
+INFO:__main__:vgg_gram_loss_5: 0.2174903005361557
+INFO:__main__:vgg_loss_0: 13.798927307128906
+INFO:__main__:vgg_loss_1: 31.710494995117188
+INFO:__main__:vgg_loss_2: 44.4327392578125
+INFO:__main__:vgg_loss_3: 33.577056884765625
+INFO:__main__:vgg_loss_4: 59.793575286865234
+INFO:__main__:vgg_loss_5: 1.992440938949585
+INFO:__main__:validation_loss: 1794.586669921875
+INFO:__main__:global_step: 18251
+INFO:__main__:kl_loss: 4720.86279296875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1839.258544921875
+INFO:__main__:loss: 1839.2633056640625
+INFO:__main__:vgg_gram_loss_0: 4.540200710296631
+INFO:__main__:vgg_gram_loss_1: 26.34616470336914
+INFO:__main__:vgg_gram_loss_2: 28.186838150024414
+INFO:__main__:vgg_gram_loss_3: 23.43221664428711
+INFO:__main__:vgg_gram_loss_4: 98.54674530029297
+INFO:__main__:vgg_gram_loss_5: 0.19403257966041565
+INFO:__main__:vgg_loss_0: 12.439112663269043
+INFO:__main__:vgg_loss_1: 31.08835220336914
+INFO:__main__:vgg_loss_2: 46.41108703613281
+INFO:__main__:vgg_loss_3: 34.44707107543945
+INFO:__main__:vgg_loss_4: 60.251800537109375
+INFO:__main__:vgg_loss_5: 1.968103289604187
+INFO:__main__:validation_loss: 2101.885009765625
+INFO:__main__:global_step: 18501
+INFO:__main__:kl_loss: 4477.91845703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1692.970947265625
+INFO:__main__:loss: 1692.9754638671875
+INFO:__main__:vgg_gram_loss_0: 12.12562084197998
+INFO:__main__:vgg_gram_loss_1: 29.67469596862793
+INFO:__main__:vgg_gram_loss_2: 27.203489303588867
+INFO:__main__:vgg_gram_loss_3: 19.88246726989746
+INFO:__main__:vgg_gram_loss_4: 83.80357360839844
+INFO:__main__:vgg_gram_loss_5: 0.18367114663124084
+INFO:__main__:vgg_loss_0: 12.472047805786133
+INFO:__main__:vgg_loss_1: 27.716249465942383
+INFO:__main__:vgg_loss_2: 40.94178009033203
+INFO:__main__:vgg_loss_3: 30.065845489501953
+INFO:__main__:vgg_loss_4: 52.750099182128906
+INFO:__main__:vgg_loss_5: 1.7746295928955078
+INFO:__main__:validation_loss: 2006.8173828125
+INFO:__main__:global_step: 18751
+INFO:__main__:kl_loss: 4860.4130859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1955.7493896484375
+INFO:__main__:loss: 1955.7542724609375
+INFO:__main__:vgg_gram_loss_0: 19.381507873535156
+INFO:__main__:vgg_gram_loss_1: 47.8715705871582
+INFO:__main__:vgg_gram_loss_2: 38.236873626708984
+INFO:__main__:vgg_gram_loss_3: 29.810638427734375
+INFO:__main__:vgg_gram_loss_4: 92.07727813720703
+INFO:__main__:vgg_gram_loss_5: 0.17494356632232666
+INFO:__main__:vgg_loss_0: 12.591361999511719
+INFO:__main__:vgg_loss_1: 31.21622657775879
+INFO:__main__:vgg_loss_2: 40.136322021484375
+INFO:__main__:vgg_loss_3: 28.3277587890625
+INFO:__main__:vgg_loss_4: 49.71849822998047
+INFO:__main__:vgg_loss_5: 1.606885552406311
+INFO:__main__:validation_loss: 2128.826904296875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 19001
+INFO:__main__:kl_loss: 4992.90576171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2126.513916015625
+INFO:__main__:loss: 2126.518798828125
+INFO:__main__:vgg_gram_loss_0: 11.762397766113281
+INFO:__main__:vgg_gram_loss_1: 35.14719009399414
+INFO:__main__:vgg_gram_loss_2: 44.32926559448242
+INFO:__main__:vgg_gram_loss_3: 34.10895919799805
+INFO:__main__:vgg_gram_loss_4: 108.45552825927734
+INFO:__main__:vgg_gram_loss_5: 0.19841814041137695
+INFO:__main__:vgg_loss_0: 13.905014991760254
+INFO:__main__:vgg_loss_1: 35.95668029785156
+INFO:__main__:vgg_loss_2: 48.03961181640625
+INFO:__main__:vgg_loss_3: 33.26726150512695
+INFO:__main__:vgg_loss_4: 58.25690841674805
+INFO:__main__:vgg_loss_5: 1.8755427598953247
+INFO:__main__:validation_loss: 1577.431640625
+INFO:__main__:global_step: 19251
+INFO:__main__:kl_loss: 5322.5244140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1775.259765625
+INFO:__main__:loss: 1775.26513671875
+INFO:__main__:vgg_gram_loss_0: 11.807829856872559
+INFO:__main__:vgg_gram_loss_1: 26.243391036987305
+INFO:__main__:vgg_gram_loss_2: 32.09434127807617
+INFO:__main__:vgg_gram_loss_3: 22.65289306640625
+INFO:__main__:vgg_gram_loss_4: 81.91146087646484
+INFO:__main__:vgg_gram_loss_5: 0.17412681877613068
+INFO:__main__:vgg_loss_0: 13.532183647155762
+INFO:__main__:vgg_loss_1: 31.96437644958496
+INFO:__main__:vgg_loss_2: 46.52936553955078
+INFO:__main__:vgg_loss_3: 32.50563049316406
+INFO:__main__:vgg_loss_4: 53.91973876953125
+INFO:__main__:vgg_loss_5: 1.7166208028793335
+INFO:__main__:validation_loss: 2093.943603515625
+INFO:__main__:global_step: 19501
+INFO:__main__:kl_loss: 4853.25146484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1649.866455078125
+INFO:__main__:loss: 1649.871337890625
+INFO:__main__:vgg_gram_loss_0: 5.752615928649902
+INFO:__main__:vgg_gram_loss_1: 19.739910125732422
+INFO:__main__:vgg_gram_loss_2: 27.144628524780273
+INFO:__main__:vgg_gram_loss_3: 19.404037475585938
+INFO:__main__:vgg_gram_loss_4: 86.3724594116211
+INFO:__main__:vgg_gram_loss_5: 0.189721018075943
+INFO:__main__:vgg_loss_0: 12.063374519348145
+INFO:__main__:vgg_loss_1: 30.16156768798828
+INFO:__main__:vgg_loss_2: 44.0325813293457
+INFO:__main__:vgg_loss_3: 30.744182586669922
+INFO:__main__:vgg_loss_4: 52.65507125854492
+INFO:__main__:vgg_loss_5: 1.7131348848342896
+INFO:__main__:validation_loss: 1981.44580078125
+INFO:__main__:global_step: 19751
+INFO:__main__:kl_loss: 5671.16796875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1668.09912109375
+INFO:__main__:loss: 1668.104736328125
+INFO:__main__:vgg_gram_loss_0: 8.434948921203613
+INFO:__main__:vgg_gram_loss_1: 28.247406005859375
+INFO:__main__:vgg_gram_loss_2: 28.897809982299805
+INFO:__main__:vgg_gram_loss_3: 22.10834312438965
+INFO:__main__:vgg_gram_loss_4: 83.86568450927734
+INFO:__main__:vgg_gram_loss_5: 0.16308985650539398
+INFO:__main__:vgg_loss_0: 12.341493606567383
+INFO:__main__:vgg_loss_1: 30.371694564819336
+INFO:__main__:vgg_loss_2: 39.89834976196289
+INFO:__main__:vgg_loss_3: 27.892358779907227
+INFO:__main__:vgg_loss_4: 49.730926513671875
+INFO:__main__:vgg_loss_5: 1.6677101850509644
+INFO:__main__:validation_loss: 1963.6214599609375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 20001
+INFO:__main__:kl_loss: 5496.36767578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1846.4300537109375
+INFO:__main__:loss: 1846.435546875
+INFO:__main__:vgg_gram_loss_0: 11.289488792419434
+INFO:__main__:vgg_gram_loss_1: 31.000329971313477
+INFO:__main__:vgg_gram_loss_2: 27.215173721313477
+INFO:__main__:vgg_gram_loss_3: 26.007368087768555
+INFO:__main__:vgg_gram_loss_4: 101.5735092163086
+INFO:__main__:vgg_gram_loss_5: 0.1830417811870575
+INFO:__main__:vgg_loss_0: 12.642966270446777
+INFO:__main__:vgg_loss_1: 30.449230194091797
+INFO:__main__:vgg_loss_2: 40.37249755859375
+INFO:__main__:vgg_loss_3: 30.449600219726562
+INFO:__main__:vgg_loss_4: 56.2236213684082
+INFO:__main__:vgg_loss_5: 1.879172444343567
+INFO:__main__:validation_loss: 1797.7470703125
+INFO:__main__:global_step: 20251
+INFO:__main__:kl_loss: 4787.83154296875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1417.0654296875
+INFO:__main__:loss: 1417.0701904296875
+INFO:__main__:vgg_gram_loss_0: 6.927775859832764
+INFO:__main__:vgg_gram_loss_1: 18.93682861328125
+INFO:__main__:vgg_gram_loss_2: 22.47786521911621
+INFO:__main__:vgg_gram_loss_3: 16.596920013427734
+INFO:__main__:vgg_gram_loss_4: 72.18505096435547
+INFO:__main__:vgg_gram_loss_5: 0.16100217401981354
+INFO:__main__:vgg_loss_0: 10.205305099487305
+INFO:__main__:vgg_loss_1: 24.550825119018555
+INFO:__main__:vgg_loss_2: 35.729862213134766
+INFO:__main__:vgg_loss_3: 26.56914520263672
+INFO:__main__:vgg_loss_4: 47.4525260925293
+INFO:__main__:vgg_loss_5: 1.6199994087219238
+INFO:__main__:validation_loss: 1864.7733154296875
+INFO:__main__:global_step: 20501
+INFO:__main__:kl_loss: 4419.68896484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1530.7275390625
+INFO:__main__:loss: 1530.73193359375
+INFO:__main__:vgg_gram_loss_0: 13.85461711883545
+INFO:__main__:vgg_gram_loss_1: 18.291284561157227
+INFO:__main__:vgg_gram_loss_2: 20.529714584350586
+INFO:__main__:vgg_gram_loss_3: 17.255401611328125
+INFO:__main__:vgg_gram_loss_4: 76.59956359863281
+INFO:__main__:vgg_gram_loss_5: 0.1520557850599289
+INFO:__main__:vgg_loss_0: 12.632413864135742
+INFO:__main__:vgg_loss_1: 27.370203018188477
+INFO:__main__:vgg_loss_2: 38.95009231567383
+INFO:__main__:vgg_loss_3: 28.210474014282227
+INFO:__main__:vgg_loss_4: 50.657588958740234
+INFO:__main__:vgg_loss_5: 1.6421043872833252
+INFO:__main__:validation_loss: 1866.90380859375
+INFO:__main__:global_step: 20751
+INFO:__main__:kl_loss: 4980.6796875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1683.9337158203125
+INFO:__main__:loss: 1683.938720703125
+INFO:__main__:vgg_gram_loss_0: 8.742308616638184
+INFO:__main__:vgg_gram_loss_1: 22.607135772705078
+INFO:__main__:vgg_gram_loss_2: 25.643049240112305
+INFO:__main__:vgg_gram_loss_3: 20.816970825195312
+INFO:__main__:vgg_gram_loss_4: 90.33566284179688
+INFO:__main__:vgg_gram_loss_5: 0.17780880630016327
+INFO:__main__:vgg_loss_0: 12.081668853759766
+INFO:__main__:vgg_loss_1: 28.6904239654541
+INFO:__main__:vgg_loss_2: 41.31591796875
+INFO:__main__:vgg_loss_3: 30.087696075439453
+INFO:__main__:vgg_loss_4: 54.46744155883789
+INFO:__main__:vgg_loss_5: 1.8206408023834229
+INFO:__main__:validation_loss: 1846.0135498046875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 21001
+INFO:__main__:kl_loss: 5041.5732421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1710.8983154296875
+INFO:__main__:loss: 1710.9033203125
+INFO:__main__:vgg_gram_loss_0: 9.825221061706543
+INFO:__main__:vgg_gram_loss_1: 24.367124557495117
+INFO:__main__:vgg_gram_loss_2: 26.96771812438965
+INFO:__main__:vgg_gram_loss_3: 20.87141990661621
+INFO:__main__:vgg_gram_loss_4: 84.66051483154297
+INFO:__main__:vgg_gram_loss_5: 0.1696486920118332
+INFO:__main__:vgg_loss_0: 12.736309051513672
+INFO:__main__:vgg_loss_1: 32.24247360229492
+INFO:__main__:vgg_loss_2: 43.70392990112305
+INFO:__main__:vgg_loss_3: 30.749725341796875
+INFO:__main__:vgg_loss_4: 54.11089324951172
+INFO:__main__:vgg_loss_5: 1.7746679782867432
+INFO:__main__:validation_loss: 2120.66259765625
+INFO:__main__:global_step: 21251
+INFO:__main__:kl_loss: 4932.83544921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1569.167724609375
+INFO:__main__:loss: 1569.172607421875
+INFO:__main__:vgg_gram_loss_0: 11.817398071289062
+INFO:__main__:vgg_gram_loss_1: 24.529876708984375
+INFO:__main__:vgg_gram_loss_2: 24.53694725036621
+INFO:__main__:vgg_gram_loss_3: 18.322492599487305
+INFO:__main__:vgg_gram_loss_4: 78.92101287841797
+INFO:__main__:vgg_gram_loss_5: 0.16936343908309937
+INFO:__main__:vgg_loss_0: 11.163140296936035
+INFO:__main__:vgg_loss_1: 26.56531524658203
+INFO:__main__:vgg_loss_2: 39.06063461303711
+INFO:__main__:vgg_loss_3: 27.787824630737305
+INFO:__main__:vgg_loss_4: 49.31822967529297
+INFO:__main__:vgg_loss_5: 1.6413142681121826
+INFO:__main__:validation_loss: 1877.9803466796875
+INFO:__main__:global_step: 21501
+INFO:__main__:kl_loss: 7461.875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1957.14111328125
+INFO:__main__:loss: 1957.1485595703125
+INFO:__main__:vgg_gram_loss_0: 7.639766216278076
+INFO:__main__:vgg_gram_loss_1: 51.336177825927734
+INFO:__main__:vgg_gram_loss_2: 27.561857223510742
+INFO:__main__:vgg_gram_loss_3: 24.805944442749023
+INFO:__main__:vgg_gram_loss_4: 93.11919403076172
+INFO:__main__:vgg_gram_loss_5: 0.1948944479227066
+INFO:__main__:vgg_loss_0: 13.733922004699707
+INFO:__main__:vgg_loss_1: 35.36043167114258
+INFO:__main__:vgg_loss_2: 44.46965026855469
+INFO:__main__:vgg_loss_3: 32.736568450927734
+INFO:__main__:vgg_loss_4: 58.56448745727539
+INFO:__main__:vgg_loss_5: 1.9053258895874023
+INFO:__main__:validation_loss: 2084.21484375
+INFO:__main__:global_step: 21751
+INFO:__main__:kl_loss: 5346.93701171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1805.1666259765625
+INFO:__main__:loss: 1805.1719970703125
+INFO:__main__:vgg_gram_loss_0: 8.519783973693848
+INFO:__main__:vgg_gram_loss_1: 23.52079200744629
+INFO:__main__:vgg_gram_loss_2: 26.342756271362305
+INFO:__main__:vgg_gram_loss_3: 21.02391242980957
+INFO:__main__:vgg_gram_loss_4: 89.01609802246094
+INFO:__main__:vgg_gram_loss_5: 0.2042589634656906
+INFO:__main__:vgg_loss_0: 14.254344940185547
+INFO:__main__:vgg_loss_1: 33.31464385986328
+INFO:__main__:vgg_loss_2: 46.98442459106445
+INFO:__main__:vgg_loss_3: 34.28253936767578
+INFO:__main__:vgg_loss_4: 61.56409454345703
+INFO:__main__:vgg_loss_5: 2.00569224357605
+INFO:__main__:validation_loss: 1630.6170654296875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 22001
+INFO:__main__:kl_loss: 7184.490234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1714.4228515625
+INFO:__main__:loss: 1714.4300537109375
+INFO:__main__:vgg_gram_loss_0: 11.607829093933105
+INFO:__main__:vgg_gram_loss_1: 33.61949157714844
+INFO:__main__:vgg_gram_loss_2: 28.496732711791992
+INFO:__main__:vgg_gram_loss_3: 22.338260650634766
+INFO:__main__:vgg_gram_loss_4: 82.47252655029297
+INFO:__main__:vgg_gram_loss_5: 0.16853296756744385
+INFO:__main__:vgg_loss_0: 12.333734512329102
+INFO:__main__:vgg_loss_1: 30.52727508544922
+INFO:__main__:vgg_loss_2: 41.495201110839844
+INFO:__main__:vgg_loss_3: 29.285791397094727
+INFO:__main__:vgg_loss_4: 48.942047119140625
+INFO:__main__:vgg_loss_5: 1.5971674919128418
+INFO:__main__:validation_loss: 2014.9185791015625
+INFO:__main__:global_step: 22251
+INFO:__main__:kl_loss: 5150.4931640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1725.57958984375
+INFO:__main__:loss: 1725.584716796875
+INFO:__main__:vgg_gram_loss_0: 13.618362426757812
+INFO:__main__:vgg_gram_loss_1: 23.64889144897461
+INFO:__main__:vgg_gram_loss_2: 25.980188369750977
+INFO:__main__:vgg_gram_loss_3: 20.49812889099121
+INFO:__main__:vgg_gram_loss_4: 86.66451263427734
+INFO:__main__:vgg_gram_loss_5: 0.1958109587430954
+INFO:__main__:vgg_loss_0: 12.967350959777832
+INFO:__main__:vgg_loss_1: 28.715744018554688
+INFO:__main__:vgg_loss_2: 41.358184814453125
+INFO:__main__:vgg_loss_3: 31.56941032409668
+INFO:__main__:vgg_loss_4: 57.91679000854492
+INFO:__main__:vgg_loss_5: 1.9824949502944946
+INFO:__main__:validation_loss: 2300.52294921875
+INFO:__main__:global_step: 22501
+INFO:__main__:kl_loss: 5312.90478515625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1798.2847900390625
+INFO:__main__:loss: 1798.2901611328125
+INFO:__main__:vgg_gram_loss_0: 9.693204879760742
+INFO:__main__:vgg_gram_loss_1: 35.04866027832031
+INFO:__main__:vgg_gram_loss_2: 30.47637939453125
+INFO:__main__:vgg_gram_loss_3: 24.25353240966797
+INFO:__main__:vgg_gram_loss_4: 81.70611572265625
+INFO:__main__:vgg_gram_loss_5: 0.18399392068386078
+INFO:__main__:vgg_loss_0: 13.380143165588379
+INFO:__main__:vgg_loss_1: 31.39232063293457
+INFO:__main__:vgg_loss_2: 43.217620849609375
+INFO:__main__:vgg_loss_3: 32.23326110839844
+INFO:__main__:vgg_loss_4: 56.16984176635742
+INFO:__main__:vgg_loss_5: 1.9018902778625488
+INFO:__main__:validation_loss: 1591.3680419921875
+INFO:__main__:global_step: 22751
+INFO:__main__:kl_loss: 4559.7900390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1466.56884765625
+INFO:__main__:loss: 1466.5733642578125
+INFO:__main__:vgg_gram_loss_0: 7.598983287811279
+INFO:__main__:vgg_gram_loss_1: 18.60493278503418
+INFO:__main__:vgg_gram_loss_2: 21.320310592651367
+INFO:__main__:vgg_gram_loss_3: 16.56488037109375
+INFO:__main__:vgg_gram_loss_4: 76.72950744628906
+INFO:__main__:vgg_gram_loss_5: 0.1628524363040924
+INFO:__main__:vgg_loss_0: 10.02210521697998
+INFO:__main__:vgg_loss_1: 23.95241355895996
+INFO:__main__:vgg_loss_2: 37.66127014160156
+INFO:__main__:vgg_loss_3: 27.937902450561523
+INFO:__main__:vgg_loss_4: 51.06528854370117
+INFO:__main__:vgg_loss_5: 1.6933501958847046
+INFO:__main__:validation_loss: 1818.0550537109375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 23001
+INFO:__main__:kl_loss: 5948.6865234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1885.263427734375
+INFO:__main__:loss: 1885.2694091796875
+INFO:__main__:vgg_gram_loss_0: 10.98624324798584
+INFO:__main__:vgg_gram_loss_1: 28.256193161010742
+INFO:__main__:vgg_gram_loss_2: 32.08843994140625
+INFO:__main__:vgg_gram_loss_3: 23.447072982788086
+INFO:__main__:vgg_gram_loss_4: 91.99577331542969
+INFO:__main__:vgg_gram_loss_5: 0.17653536796569824
+INFO:__main__:vgg_loss_0: 14.164463996887207
+INFO:__main__:vgg_loss_1: 34.57194900512695
+INFO:__main__:vgg_loss_2: 48.01591491699219
+INFO:__main__:vgg_loss_3: 33.545352935791016
+INFO:__main__:vgg_loss_4: 57.98910903930664
+INFO:__main__:vgg_loss_5: 1.8156613111495972
+INFO:__main__:validation_loss: 1723.014404296875
+INFO:__main__:global_step: 23251
+INFO:__main__:kl_loss: 5328.88671875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1721.304443359375
+INFO:__main__:loss: 1721.309814453125
+INFO:__main__:vgg_gram_loss_0: 8.487386703491211
+INFO:__main__:vgg_gram_loss_1: 25.378110885620117
+INFO:__main__:vgg_gram_loss_2: 25.46892738342285
+INFO:__main__:vgg_gram_loss_3: 19.612510681152344
+INFO:__main__:vgg_gram_loss_4: 81.29297637939453
+INFO:__main__:vgg_gram_loss_5: 0.18134182691574097
+INFO:__main__:vgg_loss_0: 14.340794563293457
+INFO:__main__:vgg_loss_1: 36.10295486450195
+INFO:__main__:vgg_loss_2: 46.48504638671875
+INFO:__main__:vgg_loss_3: 31.61625862121582
+INFO:__main__:vgg_loss_4: 53.56495666503906
+INFO:__main__:vgg_loss_5: 1.7296303510665894
+INFO:__main__:validation_loss: 2234.624755859375
+INFO:__main__:global_step: 23501
+INFO:__main__:kl_loss: 4475.861328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1489.15087890625
+INFO:__main__:loss: 1489.1553955078125
+INFO:__main__:vgg_gram_loss_0: 11.218605041503906
+INFO:__main__:vgg_gram_loss_1: 21.20235252380371
+INFO:__main__:vgg_gram_loss_2: 21.704696655273438
+INFO:__main__:vgg_gram_loss_3: 19.147504806518555
+INFO:__main__:vgg_gram_loss_4: 76.21410369873047
+INFO:__main__:vgg_gram_loss_5: 0.15340687334537506
+INFO:__main__:vgg_loss_0: 10.513860702514648
+INFO:__main__:vgg_loss_1: 25.580469131469727
+INFO:__main__:vgg_loss_2: 37.393375396728516
+INFO:__main__:vgg_loss_3: 26.532764434814453
+INFO:__main__:vgg_loss_4: 46.62006759643555
+INFO:__main__:vgg_loss_5: 1.5489575862884521
+INFO:__main__:validation_loss: 1824.110107421875
+INFO:__main__:global_step: 23751
+INFO:__main__:kl_loss: 6224.55517578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1386.58203125
+INFO:__main__:loss: 1386.5882568359375
+INFO:__main__:vgg_gram_loss_0: 7.017411708831787
+INFO:__main__:vgg_gram_loss_1: 17.303508758544922
+INFO:__main__:vgg_gram_loss_2: 18.809751510620117
+INFO:__main__:vgg_gram_loss_3: 14.905494689941406
+INFO:__main__:vgg_gram_loss_4: 71.70153045654297
+INFO:__main__:vgg_gram_loss_5: 0.1587752252817154
+INFO:__main__:vgg_loss_0: 9.687398910522461
+INFO:__main__:vgg_loss_1: 23.777997970581055
+INFO:__main__:vgg_loss_2: 36.09005355834961
+INFO:__main__:vgg_loss_3: 26.77157974243164
+INFO:__main__:vgg_loss_4: 49.452274322509766
+INFO:__main__:vgg_loss_5: 1.6406444311141968
+INFO:__main__:validation_loss: 2340.13427734375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 24001
+INFO:__main__:kl_loss: 6341.3232421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1951.650634765625
+INFO:__main__:loss: 1951.656982421875
+INFO:__main__:vgg_gram_loss_0: 8.191737174987793
+INFO:__main__:vgg_gram_loss_1: 34.70285415649414
+INFO:__main__:vgg_gram_loss_2: 38.78208923339844
+INFO:__main__:vgg_gram_loss_3: 29.2270565032959
+INFO:__main__:vgg_gram_loss_4: 97.66033935546875
+INFO:__main__:vgg_gram_loss_5: 0.1729177087545395
+INFO:__main__:vgg_loss_0: 13.707581520080566
+INFO:__main__:vgg_loss_1: 34.524784088134766
+INFO:__main__:vgg_loss_2: 46.172306060791016
+INFO:__main__:vgg_loss_3: 31.92188262939453
+INFO:__main__:vgg_loss_4: 53.5425910949707
+INFO:__main__:vgg_loss_5: 1.724003553390503
+INFO:__main__:validation_loss: 2117.68359375
+INFO:__main__:global_step: 24251
+INFO:__main__:kl_loss: 5249.95068359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1721.580810546875
+INFO:__main__:loss: 1721.5860595703125
+INFO:__main__:vgg_gram_loss_0: 10.823271751403809
+INFO:__main__:vgg_gram_loss_1: 30.82798194885254
+INFO:__main__:vgg_gram_loss_2: 27.697574615478516
+INFO:__main__:vgg_gram_loss_3: 20.705650329589844
+INFO:__main__:vgg_gram_loss_4: 85.38189697265625
+INFO:__main__:vgg_gram_loss_5: 0.16573119163513184
+INFO:__main__:vgg_loss_0: 12.905257225036621
+INFO:__main__:vgg_loss_1: 30.286806106567383
+INFO:__main__:vgg_loss_2: 41.30002212524414
+INFO:__main__:vgg_loss_3: 29.55193519592285
+INFO:__main__:vgg_loss_4: 52.948768615722656
+INFO:__main__:vgg_loss_5: 1.7212789058685303
+INFO:__main__:validation_loss: 1826.853759765625
+INFO:__main__:global_step: 24501
+INFO:__main__:kl_loss: 7116.6982421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1796.1981201171875
+INFO:__main__:loss: 1796.2052001953125
+INFO:__main__:vgg_gram_loss_0: 11.833331108093262
+INFO:__main__:vgg_gram_loss_1: 31.096662521362305
+INFO:__main__:vgg_gram_loss_2: 26.738489151000977
+INFO:__main__:vgg_gram_loss_3: 25.092031478881836
+INFO:__main__:vgg_gram_loss_4: 92.46266174316406
+INFO:__main__:vgg_gram_loss_5: 0.18547070026397705
+INFO:__main__:vgg_loss_0: 12.73943042755127
+INFO:__main__:vgg_loss_1: 29.491924285888672
+INFO:__main__:vgg_loss_2: 41.31367874145508
+INFO:__main__:vgg_loss_3: 31.352888107299805
+INFO:__main__:vgg_loss_4: 55.145469665527344
+INFO:__main__:vgg_loss_5: 1.7875674962997437
+INFO:__main__:validation_loss: 1618.7242431640625
+INFO:__main__:global_step: 24751
+INFO:__main__:kl_loss: 5999.43310546875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1486.602294921875
+INFO:__main__:loss: 1486.6082763671875
+INFO:__main__:vgg_gram_loss_0: 6.329232692718506
+INFO:__main__:vgg_gram_loss_1: 19.43773078918457
+INFO:__main__:vgg_gram_loss_2: 21.31064796447754
+INFO:__main__:vgg_gram_loss_3: 15.947096824645996
+INFO:__main__:vgg_gram_loss_4: 75.87899017333984
+INFO:__main__:vgg_gram_loss_5: 0.16657531261444092
+INFO:__main__:vgg_loss_0: 11.089919090270996
+INFO:__main__:vgg_loss_1: 25.27316665649414
+INFO:__main__:vgg_loss_2: 38.6810302734375
+INFO:__main__:vgg_loss_3: 28.890499114990234
+INFO:__main__:vgg_loss_4: 52.608612060546875
+INFO:__main__:vgg_loss_5: 1.7069624662399292
+INFO:__main__:validation_loss: 2251.744384765625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 25001
+INFO:__main__:kl_loss: 5627.7431640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1410.3660888671875
+INFO:__main__:loss: 1410.3717041015625
+INFO:__main__:vgg_gram_loss_0: 10.170222282409668
+INFO:__main__:vgg_gram_loss_1: 18.430492401123047
+INFO:__main__:vgg_gram_loss_2: 19.04314613342285
+INFO:__main__:vgg_gram_loss_3: 15.144770622253418
+INFO:__main__:vgg_gram_loss_4: 69.84618377685547
+INFO:__main__:vgg_gram_loss_5: 0.16545914113521576
+INFO:__main__:vgg_loss_0: 10.268118858337402
+INFO:__main__:vgg_loss_1: 25.2415828704834
+INFO:__main__:vgg_loss_2: 36.96037673950195
+INFO:__main__:vgg_loss_3: 26.81494903564453
+INFO:__main__:vgg_loss_4: 48.318851470947266
+INFO:__main__:vgg_loss_5: 1.6690479516983032
+INFO:__main__:validation_loss: 2013.0743408203125
+INFO:__main__:global_step: 25251
+INFO:__main__:kl_loss: 6142.14453125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1709.140869140625
+INFO:__main__:loss: 1709.14697265625
+INFO:__main__:vgg_gram_loss_0: 11.320706367492676
+INFO:__main__:vgg_gram_loss_1: 25.93552017211914
+INFO:__main__:vgg_gram_loss_2: 25.518827438354492
+INFO:__main__:vgg_gram_loss_3: 19.39504051208496
+INFO:__main__:vgg_gram_loss_4: 83.94001007080078
+INFO:__main__:vgg_gram_loss_5: 0.19050250947475433
+INFO:__main__:vgg_loss_0: 13.218282699584961
+INFO:__main__:vgg_loss_1: 28.710430145263672
+INFO:__main__:vgg_loss_2: 41.21616744995117
+INFO:__main__:vgg_loss_3: 31.4913387298584
+INFO:__main__:vgg_loss_4: 58.88178634643555
+INFO:__main__:vgg_loss_5: 2.00955867767334
+INFO:__main__:validation_loss: 1683.882080078125
+INFO:__main__:global_step: 25501
+INFO:__main__:kl_loss: 6186.7451171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1853.9647216796875
+INFO:__main__:loss: 1853.970947265625
+INFO:__main__:vgg_gram_loss_0: 20.786651611328125
+INFO:__main__:vgg_gram_loss_1: 28.144865036010742
+INFO:__main__:vgg_gram_loss_2: 27.325027465820312
+INFO:__main__:vgg_gram_loss_3: 22.583049774169922
+INFO:__main__:vgg_gram_loss_4: 92.78234100341797
+INFO:__main__:vgg_gram_loss_5: 0.18690651655197144
+INFO:__main__:vgg_loss_0: 13.260847091674805
+INFO:__main__:vgg_loss_1: 29.745174407958984
+INFO:__main__:vgg_loss_2: 43.2514533996582
+INFO:__main__:vgg_loss_3: 32.04433822631836
+INFO:__main__:vgg_loss_4: 58.720855712890625
+INFO:__main__:vgg_loss_5: 1.9614146947860718
+INFO:__main__:validation_loss: 1650.27783203125
+INFO:__main__:global_step: 25751
+INFO:__main__:kl_loss: 6973.0205078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1645.13134765625
+INFO:__main__:loss: 1645.1383056640625
+INFO:__main__:vgg_gram_loss_0: 9.280381202697754
+INFO:__main__:vgg_gram_loss_1: 27.68904685974121
+INFO:__main__:vgg_gram_loss_2: 24.16663932800293
+INFO:__main__:vgg_gram_loss_3: 21.12166404724121
+INFO:__main__:vgg_gram_loss_4: 79.82893371582031
+INFO:__main__:vgg_gram_loss_5: 0.15804514288902283
+INFO:__main__:vgg_loss_0: 12.165033340454102
+INFO:__main__:vgg_loss_1: 29.910614013671875
+INFO:__main__:vgg_loss_2: 40.40812301635742
+INFO:__main__:vgg_loss_3: 29.678409576416016
+INFO:__main__:vgg_loss_4: 52.912105560302734
+INFO:__main__:vgg_loss_5: 1.7072830200195312
+INFO:__main__:validation_loss: 2215.6083984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 26001
+INFO:__main__:kl_loss: 6218.8818359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1682.552490234375
+INFO:__main__:loss: 1682.5587158203125
+INFO:__main__:vgg_gram_loss_0: 7.330090522766113
+INFO:__main__:vgg_gram_loss_1: 23.091108322143555
+INFO:__main__:vgg_gram_loss_2: 23.08942985534668
+INFO:__main__:vgg_gram_loss_3: 19.701995849609375
+INFO:__main__:vgg_gram_loss_4: 89.66356658935547
+INFO:__main__:vgg_gram_loss_5: 0.1859285980463028
+INFO:__main__:vgg_loss_0: 13.050247192382812
+INFO:__main__:vgg_loss_1: 30.06724739074707
+INFO:__main__:vgg_loss_2: 41.38309097290039
+INFO:__main__:vgg_loss_3: 30.49485206604004
+INFO:__main__:vgg_loss_4: 56.499664306640625
+INFO:__main__:vgg_loss_5: 1.9533076286315918
+INFO:__main__:validation_loss: 1918.3074951171875
+INFO:__main__:global_step: 26251
+INFO:__main__:kl_loss: 6852.59375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1588.879638671875
+INFO:__main__:loss: 1588.886474609375
+INFO:__main__:vgg_gram_loss_0: 8.004632949829102
+INFO:__main__:vgg_gram_loss_1: 32.57810592651367
+INFO:__main__:vgg_gram_loss_2: 24.929096221923828
+INFO:__main__:vgg_gram_loss_3: 20.542936325073242
+INFO:__main__:vgg_gram_loss_4: 77.0531234741211
+INFO:__main__:vgg_gram_loss_5: 0.15604647994041443
+INFO:__main__:vgg_loss_0: 11.412250518798828
+INFO:__main__:vgg_loss_1: 27.01420021057129
+INFO:__main__:vgg_loss_2: 38.5195426940918
+INFO:__main__:vgg_loss_3: 28.043766021728516
+INFO:__main__:vgg_loss_4: 47.96466064453125
+INFO:__main__:vgg_loss_5: 1.5575368404388428
+INFO:__main__:validation_loss: 2249.79248046875
+INFO:__main__:global_step: 26501
+INFO:__main__:kl_loss: 6980.28857421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1704.5946044921875
+INFO:__main__:loss: 1704.6015625
+INFO:__main__:vgg_gram_loss_0: 12.008533477783203
+INFO:__main__:vgg_gram_loss_1: 21.613807678222656
+INFO:__main__:vgg_gram_loss_2: 26.681095123291016
+INFO:__main__:vgg_gram_loss_3: 20.401952743530273
+INFO:__main__:vgg_gram_loss_4: 84.60264587402344
+INFO:__main__:vgg_gram_loss_5: 0.1894301474094391
+INFO:__main__:vgg_loss_0: 12.829402923583984
+INFO:__main__:vgg_loss_1: 29.313905715942383
+INFO:__main__:vgg_loss_2: 43.35658264160156
+INFO:__main__:vgg_loss_3: 32.1512565612793
+INFO:__main__:vgg_loss_4: 55.89781951904297
+INFO:__main__:vgg_loss_5: 1.8724706172943115
+INFO:__main__:validation_loss: 1873.46240234375
+INFO:__main__:global_step: 26751
+INFO:__main__:kl_loss: 6226.38330078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1691.88623046875
+INFO:__main__:loss: 1691.8924560546875
+INFO:__main__:vgg_gram_loss_0: 7.683363437652588
+INFO:__main__:vgg_gram_loss_1: 21.128986358642578
+INFO:__main__:vgg_gram_loss_2: 22.99751091003418
+INFO:__main__:vgg_gram_loss_3: 19.45563316345215
+INFO:__main__:vgg_gram_loss_4: 86.91204071044922
+INFO:__main__:vgg_gram_loss_5: 0.18538795411586761
+INFO:__main__:vgg_loss_0: 13.23610782623291
+INFO:__main__:vgg_loss_1: 31.90540313720703
+INFO:__main__:vgg_loss_2: 43.89427947998047
+INFO:__main__:vgg_loss_3: 31.82846450805664
+INFO:__main__:vgg_loss_4: 57.198577880859375
+INFO:__main__:vgg_loss_5: 1.9515100717544556
+INFO:__main__:validation_loss: 2264.475830078125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 27001
+INFO:__main__:kl_loss: 6897.892578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1583.3160400390625
+INFO:__main__:loss: 1583.322998046875
+INFO:__main__:vgg_gram_loss_0: 8.693373680114746
+INFO:__main__:vgg_gram_loss_1: 21.26497459411621
+INFO:__main__:vgg_gram_loss_2: 23.890304565429688
+INFO:__main__:vgg_gram_loss_3: 17.55023765563965
+INFO:__main__:vgg_gram_loss_4: 79.44637298583984
+INFO:__main__:vgg_gram_loss_5: 0.17783643305301666
+INFO:__main__:vgg_loss_0: 11.692310333251953
+INFO:__main__:vgg_loss_1: 27.731632232666016
+INFO:__main__:vgg_loss_2: 40.75484085083008
+INFO:__main__:vgg_loss_3: 30.193740844726562
+INFO:__main__:vgg_loss_4: 53.523582458496094
+INFO:__main__:vgg_loss_5: 1.7439730167388916
+INFO:__main__:validation_loss: 1912.774169921875
+INFO:__main__:global_step: 27251
+INFO:__main__:kl_loss: 5178.7509765625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1626.1431884765625
+INFO:__main__:loss: 1626.1483154296875
+INFO:__main__:vgg_gram_loss_0: 12.652191162109375
+INFO:__main__:vgg_gram_loss_1: 32.19399642944336
+INFO:__main__:vgg_gram_loss_2: 25.745038986206055
+INFO:__main__:vgg_gram_loss_3: 21.77153778076172
+INFO:__main__:vgg_gram_loss_4: 80.12590789794922
+INFO:__main__:vgg_gram_loss_5: 0.1634579300880432
+INFO:__main__:vgg_loss_0: 11.062541007995605
+INFO:__main__:vgg_loss_1: 27.467317581176758
+INFO:__main__:vgg_loss_2: 38.08487319946289
+INFO:__main__:vgg_loss_3: 27.266159057617188
+INFO:__main__:vgg_loss_4: 47.11213684082031
+INFO:__main__:vgg_loss_5: 1.5835005044937134
+INFO:__main__:validation_loss: 1783.89697265625
+INFO:__main__:global_step: 27501
+INFO:__main__:kl_loss: 6521.7353515625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1723.310302734375
+INFO:__main__:loss: 1723.3167724609375
+INFO:__main__:vgg_gram_loss_0: 11.551633834838867
+INFO:__main__:vgg_gram_loss_1: 26.776636123657227
+INFO:__main__:vgg_gram_loss_2: 25.140085220336914
+INFO:__main__:vgg_gram_loss_3: 19.369688034057617
+INFO:__main__:vgg_gram_loss_4: 83.29908752441406
+INFO:__main__:vgg_gram_loss_5: 0.19278185069561005
+INFO:__main__:vgg_loss_0: 12.514986038208008
+INFO:__main__:vgg_loss_1: 29.113794326782227
+INFO:__main__:vgg_loss_2: 44.04909896850586
+INFO:__main__:vgg_loss_3: 32.55424118041992
+INFO:__main__:vgg_loss_4: 58.1593017578125
+INFO:__main__:vgg_loss_5: 1.9407258033752441
+INFO:__main__:validation_loss: 1973.332275390625
+INFO:__main__:global_step: 27751
+INFO:__main__:kl_loss: 7555.8271484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1700.685791015625
+INFO:__main__:loss: 1700.693359375
+INFO:__main__:vgg_gram_loss_0: 10.452322959899902
+INFO:__main__:vgg_gram_loss_1: 24.14700698852539
+INFO:__main__:vgg_gram_loss_2: 26.62457275390625
+INFO:__main__:vgg_gram_loss_3: 19.78400421142578
+INFO:__main__:vgg_gram_loss_4: 84.8041000366211
+INFO:__main__:vgg_gram_loss_5: 0.1840541511774063
+INFO:__main__:vgg_loss_0: 12.366714477539062
+INFO:__main__:vgg_loss_1: 31.705739974975586
+INFO:__main__:vgg_loss_2: 42.88370895385742
+INFO:__main__:vgg_loss_3: 30.60978126525879
+INFO:__main__:vgg_loss_4: 54.733646392822266
+INFO:__main__:vgg_loss_5: 1.8414698839187622
+INFO:__main__:validation_loss: 1748.7911376953125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 28001
+INFO:__main__:kl_loss: 7006.484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1801.0430908203125
+INFO:__main__:loss: 1801.050048828125
+INFO:__main__:vgg_gram_loss_0: 13.374608039855957
+INFO:__main__:vgg_gram_loss_1: 22.997451782226562
+INFO:__main__:vgg_gram_loss_2: 26.51910400390625
+INFO:__main__:vgg_gram_loss_3: 21.808897018432617
+INFO:__main__:vgg_gram_loss_4: 88.30428314208984
+INFO:__main__:vgg_gram_loss_5: 0.18019089102745056
+INFO:__main__:vgg_loss_0: 14.381964683532715
+INFO:__main__:vgg_loss_1: 34.07027053833008
+INFO:__main__:vgg_loss_2: 46.85667037963867
+INFO:__main__:vgg_loss_3: 32.78913497924805
+INFO:__main__:vgg_loss_4: 57.058799743652344
+INFO:__main__:vgg_loss_5: 1.8672504425048828
+INFO:__main__:validation_loss: 2233.9267578125
+INFO:__main__:global_step: 28251
+INFO:__main__:kl_loss: 6991.6796875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1570.6463623046875
+INFO:__main__:loss: 1570.6533203125
+INFO:__main__:vgg_gram_loss_0: 8.678009986877441
+INFO:__main__:vgg_gram_loss_1: 25.753881454467773
+INFO:__main__:vgg_gram_loss_2: 21.553869247436523
+INFO:__main__:vgg_gram_loss_3: 17.74287986755371
+INFO:__main__:vgg_gram_loss_4: 75.90467834472656
+INFO:__main__:vgg_gram_loss_5: 0.15736109018325806
+INFO:__main__:vgg_loss_0: 12.255733489990234
+INFO:__main__:vgg_loss_1: 30.231689453125
+INFO:__main__:vgg_loss_2: 40.39585876464844
+INFO:__main__:vgg_loss_3: 29.1583251953125
+INFO:__main__:vgg_loss_4: 50.635379791259766
+INFO:__main__:vgg_loss_5: 1.6616016626358032
+INFO:__main__:validation_loss: 2021.83544921875
+INFO:__main__:global_step: 28501
+INFO:__main__:kl_loss: 7319.66455078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1507.8668212890625
+INFO:__main__:loss: 1507.8741455078125
+INFO:__main__:vgg_gram_loss_0: 13.68748664855957
+INFO:__main__:vgg_gram_loss_1: 29.370996475219727
+INFO:__main__:vgg_gram_loss_2: 24.294782638549805
+INFO:__main__:vgg_gram_loss_3: 18.50814437866211
+INFO:__main__:vgg_gram_loss_4: 72.04717254638672
+INFO:__main__:vgg_gram_loss_5: 0.15147070586681366
+INFO:__main__:vgg_loss_0: 11.467558860778809
+INFO:__main__:vgg_loss_1: 27.06207275390625
+INFO:__main__:vgg_loss_2: 34.5655517578125
+INFO:__main__:vgg_loss_3: 25.13874626159668
+INFO:__main__:vgg_loss_4: 43.76496505737305
+INFO:__main__:vgg_loss_5: 1.5144022703170776
+INFO:__main__:validation_loss: 2028.8160400390625
+INFO:__main__:global_step: 28751
+INFO:__main__:kl_loss: 6046.599609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1541.998291015625
+INFO:__main__:loss: 1542.00439453125
+INFO:__main__:vgg_gram_loss_0: 13.567819595336914
+INFO:__main__:vgg_gram_loss_1: 24.398635864257812
+INFO:__main__:vgg_gram_loss_2: 21.604875564575195
+INFO:__main__:vgg_gram_loss_3: 18.050750732421875
+INFO:__main__:vgg_gram_loss_4: 80.14633178710938
+INFO:__main__:vgg_gram_loss_5: 0.15791325271129608
+INFO:__main__:vgg_loss_0: 11.199384689331055
+INFO:__main__:vgg_loss_1: 26.727035522460938
+INFO:__main__:vgg_loss_2: 37.31371307373047
+INFO:__main__:vgg_loss_3: 26.51299285888672
+INFO:__main__:vgg_loss_4: 47.15300369262695
+INFO:__main__:vgg_loss_5: 1.5671954154968262
+INFO:__main__:validation_loss: 1630.343505859375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 29001
+INFO:__main__:kl_loss: 6740.71826171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1682.2861328125
+INFO:__main__:loss: 1682.2928466796875
+INFO:__main__:vgg_gram_loss_0: 8.695304870605469
+INFO:__main__:vgg_gram_loss_1: 19.400619506835938
+INFO:__main__:vgg_gram_loss_2: 22.67316436767578
+INFO:__main__:vgg_gram_loss_3: 20.046979904174805
+INFO:__main__:vgg_gram_loss_4: 87.79853820800781
+INFO:__main__:vgg_gram_loss_5: 0.19452379643917084
+INFO:__main__:vgg_loss_0: 12.044966697692871
+INFO:__main__:vgg_loss_1: 29.201589584350586
+INFO:__main__:vgg_loss_2: 44.07809066772461
+INFO:__main__:vgg_loss_3: 32.421363830566406
+INFO:__main__:vgg_loss_4: 57.96095657348633
+INFO:__main__:vgg_loss_5: 1.9411317110061646
+INFO:__main__:validation_loss: 1976.0982666015625
+INFO:__main__:global_step: 29251
+INFO:__main__:kl_loss: 6531.47607421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1465.192626953125
+INFO:__main__:loss: 1465.19921875
+INFO:__main__:vgg_gram_loss_0: 5.189122676849365
+INFO:__main__:vgg_gram_loss_1: 20.3861141204834
+INFO:__main__:vgg_gram_loss_2: 20.574844360351562
+INFO:__main__:vgg_gram_loss_3: 17.603164672851562
+INFO:__main__:vgg_gram_loss_4: 77.5077896118164
+INFO:__main__:vgg_gram_loss_5: 0.16504782438278198
+INFO:__main__:vgg_loss_0: 10.536651611328125
+INFO:__main__:vgg_loss_1: 25.29227066040039
+INFO:__main__:vgg_loss_2: 38.12633514404297
+INFO:__main__:vgg_loss_3: 27.339921951293945
+INFO:__main__:vgg_loss_4: 48.62360763549805
+INFO:__main__:vgg_loss_5: 1.69362211227417
+INFO:__main__:validation_loss: 2093.66845703125
+INFO:__main__:global_step: 29501
+INFO:__main__:kl_loss: 7027.560546875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1550.742431640625
+INFO:__main__:loss: 1550.74951171875
+INFO:__main__:vgg_gram_loss_0: 9.524287223815918
+INFO:__main__:vgg_gram_loss_1: 18.086713790893555
+INFO:__main__:vgg_gram_loss_2: 21.886558532714844
+INFO:__main__:vgg_gram_loss_3: 18.569007873535156
+INFO:__main__:vgg_gram_loss_4: 83.06659698486328
+INFO:__main__:vgg_gram_loss_5: 0.17091219127178192
+INFO:__main__:vgg_loss_0: 10.751038551330566
+INFO:__main__:vgg_loss_1: 24.940603256225586
+INFO:__main__:vgg_loss_2: 39.313690185546875
+INFO:__main__:vgg_loss_3: 29.500207901000977
+INFO:__main__:vgg_loss_4: 52.622894287109375
+INFO:__main__:vgg_loss_5: 1.7159700393676758
+INFO:__main__:validation_loss: 1794.6380615234375
+INFO:__main__:global_step: 29751
+INFO:__main__:kl_loss: 8569.4365234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1732.98583984375
+INFO:__main__:loss: 1732.994384765625
+INFO:__main__:vgg_gram_loss_0: 12.27071762084961
+INFO:__main__:vgg_gram_loss_1: 24.522642135620117
+INFO:__main__:vgg_gram_loss_2: 29.224720001220703
+INFO:__main__:vgg_gram_loss_3: 20.72979736328125
+INFO:__main__:vgg_gram_loss_4: 79.61004638671875
+INFO:__main__:vgg_gram_loss_5: 0.17849066853523254
+INFO:__main__:vgg_loss_0: 13.388283729553223
+INFO:__main__:vgg_loss_1: 31.589082717895508
+INFO:__main__:vgg_loss_2: 46.221923828125
+INFO:__main__:vgg_loss_3: 32.76070022583008
+INFO:__main__:vgg_loss_4: 54.363521575927734
+INFO:__main__:vgg_loss_5: 1.7372430562973022
+INFO:__main__:validation_loss: 2095.364501953125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 30001
+INFO:__main__:kl_loss: 7812.275390625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1689.72021484375
+INFO:__main__:loss: 1689.72802734375
+INFO:__main__:vgg_gram_loss_0: 7.3410797119140625
+INFO:__main__:vgg_gram_loss_1: 22.4700984954834
+INFO:__main__:vgg_gram_loss_2: 21.892820358276367
+INFO:__main__:vgg_gram_loss_3: 18.213821411132812
+INFO:__main__:vgg_gram_loss_4: 85.97173309326172
+INFO:__main__:vgg_gram_loss_5: 0.19198797643184662
+INFO:__main__:vgg_loss_0: 12.74738597869873
+INFO:__main__:vgg_loss_1: 30.815444946289062
+INFO:__main__:vgg_loss_2: 44.6407470703125
+INFO:__main__:vgg_loss_3: 32.8369026184082
+INFO:__main__:vgg_loss_4: 58.828250885009766
+INFO:__main__:vgg_loss_5: 1.993740439414978
+INFO:__main__:validation_loss: 2419.019287109375
+INFO:__main__:global_step: 30251
+INFO:__main__:kl_loss: 8521.3984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1964.75830078125
+INFO:__main__:loss: 1964.766845703125
+INFO:__main__:vgg_gram_loss_0: 6.659420967102051
+INFO:__main__:vgg_gram_loss_1: 22.764286041259766
+INFO:__main__:vgg_gram_loss_2: 33.33052444458008
+INFO:__main__:vgg_gram_loss_3: 27.01814079284668
+INFO:__main__:vgg_gram_loss_4: 101.89287567138672
+INFO:__main__:vgg_gram_loss_5: 0.21862566471099854
+INFO:__main__:vgg_loss_0: 15.357515335083008
+INFO:__main__:vgg_loss_1: 34.13472366333008
+INFO:__main__:vgg_loss_2: 50.27109146118164
+INFO:__main__:vgg_loss_3: 36.350791931152344
+INFO:__main__:vgg_loss_4: 62.87358093261719
+INFO:__main__:vgg_loss_5: 2.0800621509552
+INFO:__main__:validation_loss: 2112.103515625
+INFO:__main__:global_step: 30501
+INFO:__main__:kl_loss: 7397.51220703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1605.8050537109375
+INFO:__main__:loss: 1605.8125
+INFO:__main__:vgg_gram_loss_0: 8.002279281616211
+INFO:__main__:vgg_gram_loss_1: 23.73541831970215
+INFO:__main__:vgg_gram_loss_2: 24.308568954467773
+INFO:__main__:vgg_gram_loss_3: 19.924631118774414
+INFO:__main__:vgg_gram_loss_4: 80.81922912597656
+INFO:__main__:vgg_gram_loss_5: 0.16787424683570862
+INFO:__main__:vgg_loss_0: 11.824690818786621
+INFO:__main__:vgg_loss_1: 29.90940284729004
+INFO:__main__:vgg_loss_2: 42.416412353515625
+INFO:__main__:vgg_loss_3: 28.987524032592773
+INFO:__main__:vgg_loss_4: 49.466102600097656
+INFO:__main__:vgg_loss_5: 1.5988966226577759
+INFO:__main__:validation_loss: 1885.314208984375
+INFO:__main__:global_step: 30751
+INFO:__main__:kl_loss: 8702.7744140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1596.2640380859375
+INFO:__main__:loss: 1596.272705078125
+INFO:__main__:vgg_gram_loss_0: 4.084274768829346
+INFO:__main__:vgg_gram_loss_1: 18.62978744506836
+INFO:__main__:vgg_gram_loss_2: 21.505178451538086
+INFO:__main__:vgg_gram_loss_3: 17.89800453186035
+INFO:__main__:vgg_gram_loss_4: 81.99454498291016
+INFO:__main__:vgg_gram_loss_5: 0.19694426655769348
+INFO:__main__:vgg_loss_0: 11.946273803710938
+INFO:__main__:vgg_loss_1: 29.416860580444336
+INFO:__main__:vgg_loss_2: 43.647804260253906
+INFO:__main__:vgg_loss_3: 31.734018325805664
+INFO:__main__:vgg_loss_4: 56.293212890625
+INFO:__main__:vgg_loss_5: 1.9059001207351685
+INFO:__main__:validation_loss: 1580.9931640625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 31001
+INFO:__main__:kl_loss: 7422.7265625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1665.5206298828125
+INFO:__main__:loss: 1665.528076171875
+INFO:__main__:vgg_gram_loss_0: 7.4895195960998535
+INFO:__main__:vgg_gram_loss_1: 20.219242095947266
+INFO:__main__:vgg_gram_loss_2: 27.897180557250977
+INFO:__main__:vgg_gram_loss_3: 19.956405639648438
+INFO:__main__:vgg_gram_loss_4: 85.25379180908203
+INFO:__main__:vgg_gram_loss_5: 0.1990671306848526
+INFO:__main__:vgg_loss_0: 12.088481903076172
+INFO:__main__:vgg_loss_1: 29.120532989501953
+INFO:__main__:vgg_loss_2: 42.53718566894531
+INFO:__main__:vgg_loss_3: 31.43267059326172
+INFO:__main__:vgg_loss_4: 55.092994689941406
+INFO:__main__:vgg_loss_5: 1.8170398473739624
+INFO:__main__:validation_loss: 1984.239501953125
+INFO:__main__:global_step: 31251
+INFO:__main__:kl_loss: 8185.1015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1604.841552734375
+INFO:__main__:loss: 1604.8497314453125
+INFO:__main__:vgg_gram_loss_0: 13.089444160461426
+INFO:__main__:vgg_gram_loss_1: 22.364084243774414
+INFO:__main__:vgg_gram_loss_2: 21.805578231811523
+INFO:__main__:vgg_gram_loss_3: 18.109394073486328
+INFO:__main__:vgg_gram_loss_4: 77.81670379638672
+INFO:__main__:vgg_gram_loss_5: 0.17303717136383057
+INFO:__main__:vgg_loss_0: 13.159098625183105
+INFO:__main__:vgg_loss_1: 28.713388442993164
+INFO:__main__:vgg_loss_2: 40.76849365234375
+INFO:__main__:vgg_loss_3: 30.154808044433594
+INFO:__main__:vgg_loss_4: 53.038516998291016
+INFO:__main__:vgg_loss_5: 1.7757582664489746
+INFO:__main__:validation_loss: 1555.4007568359375
+INFO:__main__:global_step: 31501
+INFO:__main__:kl_loss: 7402.59375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1490.0303955078125
+INFO:__main__:loss: 1490.037841796875
+INFO:__main__:vgg_gram_loss_0: 9.725199699401855
+INFO:__main__:vgg_gram_loss_1: 23.177865982055664
+INFO:__main__:vgg_gram_loss_2: 25.40339469909668
+INFO:__main__:vgg_gram_loss_3: 18.384431838989258
+INFO:__main__:vgg_gram_loss_4: 75.72759246826172
+INFO:__main__:vgg_gram_loss_5: 0.1574365347623825
+INFO:__main__:vgg_loss_0: 9.718292236328125
+INFO:__main__:vgg_loss_1: 23.763614654541016
+INFO:__main__:vgg_loss_2: 36.29725646972656
+INFO:__main__:vgg_loss_3: 26.906002044677734
+INFO:__main__:vgg_loss_4: 47.20603561401367
+INFO:__main__:vgg_loss_5: 1.5389440059661865
+INFO:__main__:validation_loss: 1775.9581298828125
+INFO:__main__:global_step: 31751
+INFO:__main__:kl_loss: 8372.71875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1771.30517578125
+INFO:__main__:loss: 1771.3135986328125
+INFO:__main__:vgg_gram_loss_0: 12.17890453338623
+INFO:__main__:vgg_gram_loss_1: 26.358612060546875
+INFO:__main__:vgg_gram_loss_2: 30.13270378112793
+INFO:__main__:vgg_gram_loss_3: 23.5117244720459
+INFO:__main__:vgg_gram_loss_4: 88.96038055419922
+INFO:__main__:vgg_gram_loss_5: 0.18301668763160706
+INFO:__main__:vgg_loss_0: 12.808883666992188
+INFO:__main__:vgg_loss_1: 30.362239837646484
+INFO:__main__:vgg_loss_2: 42.860774993896484
+INFO:__main__:vgg_loss_3: 31.0824031829834
+INFO:__main__:vgg_loss_4: 54.00221252441406
+INFO:__main__:vgg_loss_5: 1.8191654682159424
+INFO:__main__:validation_loss: 2132.423828125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 32001
+INFO:__main__:kl_loss: 7507.0712890625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1368.962646484375
+INFO:__main__:loss: 1368.9700927734375
+INFO:__main__:vgg_gram_loss_0: 7.300660133361816
+INFO:__main__:vgg_gram_loss_1: 15.08447551727295
+INFO:__main__:vgg_gram_loss_2: 19.11664390563965
+INFO:__main__:vgg_gram_loss_3: 15.4403657913208
+INFO:__main__:vgg_gram_loss_4: 71.96074676513672
+INFO:__main__:vgg_gram_loss_5: 0.15559887886047363
+INFO:__main__:vgg_loss_0: 9.949869155883789
+INFO:__main__:vgg_loss_1: 23.242774963378906
+INFO:__main__:vgg_loss_2: 35.481956481933594
+INFO:__main__:vgg_loss_3: 26.68964385986328
+INFO:__main__:vgg_loss_4: 47.822391510009766
+INFO:__main__:vgg_loss_5: 1.547431468963623
+INFO:__main__:validation_loss: 1999.1343994140625
+INFO:__main__:global_step: 32251
+INFO:__main__:kl_loss: 8140.81494140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1863.8477783203125
+INFO:__main__:loss: 1863.85595703125
+INFO:__main__:vgg_gram_loss_0: 6.013930797576904
+INFO:__main__:vgg_gram_loss_1: 35.12582778930664
+INFO:__main__:vgg_gram_loss_2: 34.75369644165039
+INFO:__main__:vgg_gram_loss_3: 28.84720230102539
+INFO:__main__:vgg_gram_loss_4: 87.96319580078125
+INFO:__main__:vgg_gram_loss_5: 0.1937442272901535
+INFO:__main__:vgg_loss_0: 13.145959854125977
+INFO:__main__:vgg_loss_1: 32.48296356201172
+INFO:__main__:vgg_loss_2: 45.67557907104492
+INFO:__main__:vgg_loss_3: 32.40920639038086
+INFO:__main__:vgg_loss_4: 54.372718811035156
+INFO:__main__:vgg_loss_5: 1.7855448722839355
+INFO:__main__:validation_loss: 2302.90673828125
+INFO:__main__:global_step: 32501
+INFO:__main__:kl_loss: 7173.57470703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1423.6370849609375
+INFO:__main__:loss: 1423.644287109375
+INFO:__main__:vgg_gram_loss_0: 10.145112991333008
+INFO:__main__:vgg_gram_loss_1: 21.457168579101562
+INFO:__main__:vgg_gram_loss_2: 20.230438232421875
+INFO:__main__:vgg_gram_loss_3: 16.04782485961914
+INFO:__main__:vgg_gram_loss_4: 72.99710083007812
+INFO:__main__:vgg_gram_loss_5: 0.1581924557685852
+INFO:__main__:vgg_loss_0: 11.063898086547852
+INFO:__main__:vgg_loss_1: 24.61655616760254
+INFO:__main__:vgg_loss_2: 34.753379821777344
+INFO:__main__:vgg_loss_3: 25.293195724487305
+INFO:__main__:vgg_loss_4: 46.3483772277832
+INFO:__main__:vgg_loss_5: 1.6162017583847046
+INFO:__main__:validation_loss: 1436.0740966796875
+INFO:__main__:global_step: 32751
+INFO:__main__:kl_loss: 6524.986328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1567.75244140625
+INFO:__main__:loss: 1567.7589111328125
+INFO:__main__:vgg_gram_loss_0: 15.303736686706543
+INFO:__main__:vgg_gram_loss_1: 23.70351219177246
+INFO:__main__:vgg_gram_loss_2: 26.473554611206055
+INFO:__main__:vgg_gram_loss_3: 19.5965633392334
+INFO:__main__:vgg_gram_loss_4: 78.61532592773438
+INFO:__main__:vgg_gram_loss_5: 0.15331611037254333
+INFO:__main__:vgg_loss_0: 11.39196491241455
+INFO:__main__:vgg_loss_1: 27.400014877319336
+INFO:__main__:vgg_loss_2: 37.67536544799805
+INFO:__main__:vgg_loss_3: 25.966724395751953
+INFO:__main__:vgg_loss_4: 45.72749328613281
+INFO:__main__:vgg_loss_5: 1.5428931713104248
+INFO:__main__:validation_loss: 2331.234130859375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 33001
+INFO:__main__:kl_loss: 10939.5009765625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1900.011962890625
+INFO:__main__:loss: 1900.02294921875
+INFO:__main__:vgg_gram_loss_0: 7.07985258102417
+INFO:__main__:vgg_gram_loss_1: 27.271387100219727
+INFO:__main__:vgg_gram_loss_2: 35.25288391113281
+INFO:__main__:vgg_gram_loss_3: 30.127553939819336
+INFO:__main__:vgg_gram_loss_4: 99.07093048095703
+INFO:__main__:vgg_gram_loss_5: 0.1930004358291626
+INFO:__main__:vgg_loss_0: 13.147265434265137
+INFO:__main__:vgg_loss_1: 33.134429931640625
+INFO:__main__:vgg_loss_2: 46.22199630737305
+INFO:__main__:vgg_loss_3: 32.368289947509766
+INFO:__main__:vgg_loss_4: 54.38119888305664
+INFO:__main__:vgg_loss_5: 1.75357985496521
+INFO:__main__:validation_loss: 1508.9962158203125
+INFO:__main__:global_step: 33251
+INFO:__main__:kl_loss: 7553.2265625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1798.7579345703125
+INFO:__main__:loss: 1798.7655029296875
+INFO:__main__:vgg_gram_loss_0: 12.825403213500977
+INFO:__main__:vgg_gram_loss_1: 29.995691299438477
+INFO:__main__:vgg_gram_loss_2: 26.814319610595703
+INFO:__main__:vgg_gram_loss_3: 23.321575164794922
+INFO:__main__:vgg_gram_loss_4: 88.9933090209961
+INFO:__main__:vgg_gram_loss_5: 0.1796519160270691
+INFO:__main__:vgg_loss_0: 13.508710861206055
+INFO:__main__:vgg_loss_1: 34.84256362915039
+INFO:__main__:vgg_loss_2: 42.6619758605957
+INFO:__main__:vgg_loss_3: 30.537158966064453
+INFO:__main__:vgg_loss_4: 54.273677825927734
+INFO:__main__:vgg_loss_5: 1.7975609302520752
+INFO:__main__:validation_loss: 2110.814208984375
+INFO:__main__:global_step: 33501
+INFO:__main__:kl_loss: 7539.8232421875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1470.26708984375
+INFO:__main__:loss: 1470.274658203125
+INFO:__main__:vgg_gram_loss_0: 7.521893501281738
+INFO:__main__:vgg_gram_loss_1: 18.998716354370117
+INFO:__main__:vgg_gram_loss_2: 20.80803108215332
+INFO:__main__:vgg_gram_loss_3: 17.182662963867188
+INFO:__main__:vgg_gram_loss_4: 73.57921600341797
+INFO:__main__:vgg_gram_loss_5: 0.15982714295387268
+INFO:__main__:vgg_loss_0: 11.317607879638672
+INFO:__main__:vgg_loss_1: 27.017763137817383
+INFO:__main__:vgg_loss_2: 38.52604293823242
+INFO:__main__:vgg_loss_3: 28.11553382873535
+INFO:__main__:vgg_loss_4: 49.25281524658203
+INFO:__main__:vgg_loss_5: 1.5733047723770142
+INFO:__main__:validation_loss: 1999.73974609375
+INFO:__main__:global_step: 33751
+INFO:__main__:kl_loss: 6667.193359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1280.2393798828125
+INFO:__main__:loss: 1280.24609375
+INFO:__main__:vgg_gram_loss_0: 8.491766929626465
+INFO:__main__:vgg_gram_loss_1: 16.49424171447754
+INFO:__main__:vgg_gram_loss_2: 17.525793075561523
+INFO:__main__:vgg_gram_loss_3: 13.442642211914062
+INFO:__main__:vgg_gram_loss_4: 63.419002532958984
+INFO:__main__:vgg_gram_loss_5: 0.14162833988666534
+INFO:__main__:vgg_loss_0: 9.988456726074219
+INFO:__main__:vgg_loss_1: 24.418048858642578
+INFO:__main__:vgg_loss_2: 34.24618148803711
+INFO:__main__:vgg_loss_3: 24.316545486450195
+INFO:__main__:vgg_loss_4: 42.19325637817383
+INFO:__main__:vgg_loss_5: 1.370316743850708
+INFO:__main__:validation_loss: 1770.0706787109375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 34001
+INFO:__main__:kl_loss: 7476.705078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1807.9027099609375
+INFO:__main__:loss: 1807.91015625
+INFO:__main__:vgg_gram_loss_0: 9.691965103149414
+INFO:__main__:vgg_gram_loss_1: 25.24032211303711
+INFO:__main__:vgg_gram_loss_2: 31.78862953186035
+INFO:__main__:vgg_gram_loss_3: 23.491683959960938
+INFO:__main__:vgg_gram_loss_4: 90.4525375366211
+INFO:__main__:vgg_gram_loss_5: 0.2146189957857132
+INFO:__main__:vgg_loss_0: 12.932808876037598
+INFO:__main__:vgg_loss_1: 32.66667556762695
+INFO:__main__:vgg_loss_2: 46.28645706176758
+INFO:__main__:vgg_loss_3: 32.0604248046875
+INFO:__main__:vgg_loss_4: 54.852962493896484
+INFO:__main__:vgg_loss_5: 1.9014643430709839
+INFO:__main__:validation_loss: 1683.9722900390625
+INFO:__main__:global_step: 34251
+INFO:__main__:kl_loss: 7209.8486328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1582.0489501953125
+INFO:__main__:loss: 1582.05615234375
+INFO:__main__:vgg_gram_loss_0: 7.021142482757568
+INFO:__main__:vgg_gram_loss_1: 23.712890625
+INFO:__main__:vgg_gram_loss_2: 22.4034423828125
+INFO:__main__:vgg_gram_loss_3: 19.306621551513672
+INFO:__main__:vgg_gram_loss_4: 83.2828598022461
+INFO:__main__:vgg_gram_loss_5: 0.16146491467952728
+INFO:__main__:vgg_loss_0: 11.05384349822998
+INFO:__main__:vgg_loss_1: 27.824371337890625
+INFO:__main__:vgg_loss_2: 40.05014419555664
+INFO:__main__:vgg_loss_3: 28.684080123901367
+INFO:__main__:vgg_loss_4: 51.2217903137207
+INFO:__main__:vgg_loss_5: 1.6871557235717773
+INFO:__main__:validation_loss: 1809.0576171875
+INFO:__main__:global_step: 34501
+INFO:__main__:kl_loss: 8045.5849609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1911.607666015625
+INFO:__main__:loss: 1911.61572265625
+INFO:__main__:vgg_gram_loss_0: 7.64352560043335
+INFO:__main__:vgg_gram_loss_1: 30.03386878967285
+INFO:__main__:vgg_gram_loss_2: 35.1494026184082
+INFO:__main__:vgg_gram_loss_3: 27.489959716796875
+INFO:__main__:vgg_gram_loss_4: 92.80669403076172
+INFO:__main__:vgg_gram_loss_5: 0.1786065399646759
+INFO:__main__:vgg_loss_0: 13.750406265258789
+INFO:__main__:vgg_loss_1: 35.86495590209961
+INFO:__main__:vgg_loss_2: 48.537879943847656
+INFO:__main__:vgg_loss_3: 33.572113037109375
+INFO:__main__:vgg_loss_4: 55.488426208496094
+INFO:__main__:vgg_loss_5: 1.8057047128677368
+INFO:__main__:validation_loss: 1884.7188720703125
+INFO:__main__:global_step: 34751
+INFO:__main__:kl_loss: 8501.81640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1621.1163330078125
+INFO:__main__:loss: 1621.1248779296875
+INFO:__main__:vgg_gram_loss_0: 8.063578605651855
+INFO:__main__:vgg_gram_loss_1: 27.075748443603516
+INFO:__main__:vgg_gram_loss_2: 23.694639205932617
+INFO:__main__:vgg_gram_loss_3: 19.012617111206055
+INFO:__main__:vgg_gram_loss_4: 76.67565155029297
+INFO:__main__:vgg_gram_loss_5: 0.16126571595668793
+INFO:__main__:vgg_loss_0: 13.592759132385254
+INFO:__main__:vgg_loss_1: 31.755699157714844
+INFO:__main__:vgg_loss_2: 41.26490020751953
+INFO:__main__:vgg_loss_3: 29.42359733581543
+INFO:__main__:vgg_loss_4: 51.78695297241211
+INFO:__main__:vgg_loss_5: 1.7158440351486206
+INFO:__main__:validation_loss: 2097.615478515625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 35001
+INFO:__main__:kl_loss: 7987.68798828125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1503.52880859375
+INFO:__main__:loss: 1503.5367431640625
+INFO:__main__:vgg_gram_loss_0: 9.600152969360352
+INFO:__main__:vgg_gram_loss_1: 18.655702590942383
+INFO:__main__:vgg_gram_loss_2: 18.9547176361084
+INFO:__main__:vgg_gram_loss_3: 16.299245834350586
+INFO:__main__:vgg_gram_loss_4: 74.2104721069336
+INFO:__main__:vgg_gram_loss_5: 0.18371054530143738
+INFO:__main__:vgg_loss_0: 12.533797264099121
+INFO:__main__:vgg_loss_1: 29.307140350341797
+INFO:__main__:vgg_loss_2: 40.745513916015625
+INFO:__main__:vgg_loss_3: 28.7817325592041
+INFO:__main__:vgg_loss_4: 49.73834991455078
+INFO:__main__:vgg_loss_5: 1.6951854228973389
+INFO:__main__:validation_loss: 1906.821044921875
+INFO:__main__:global_step: 35251
+INFO:__main__:kl_loss: 8501.720703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1648.5947265625
+INFO:__main__:loss: 1648.603271484375
+INFO:__main__:vgg_gram_loss_0: 10.086333274841309
+INFO:__main__:vgg_gram_loss_1: 22.291767120361328
+INFO:__main__:vgg_gram_loss_2: 22.4310359954834
+INFO:__main__:vgg_gram_loss_3: 19.004484176635742
+INFO:__main__:vgg_gram_loss_4: 89.95596313476562
+INFO:__main__:vgg_gram_loss_5: 0.18401210010051727
+INFO:__main__:vgg_loss_0: 11.595192909240723
+INFO:__main__:vgg_loss_1: 27.05489730834961
+INFO:__main__:vgg_loss_2: 40.13462448120117
+INFO:__main__:vgg_loss_3: 30.07659339904785
+INFO:__main__:vgg_loss_4: 55.03064727783203
+INFO:__main__:vgg_loss_5: 1.8733744621276855
+INFO:__main__:validation_loss: 1783.89013671875
+INFO:__main__:global_step: 35501
+INFO:__main__:kl_loss: 8287.830078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1577.493896484375
+INFO:__main__:loss: 1577.502197265625
+INFO:__main__:vgg_gram_loss_0: 10.192741394042969
+INFO:__main__:vgg_gram_loss_1: 20.201032638549805
+INFO:__main__:vgg_gram_loss_2: 21.043058395385742
+INFO:__main__:vgg_gram_loss_3: 17.567325592041016
+INFO:__main__:vgg_gram_loss_4: 82.69979095458984
+INFO:__main__:vgg_gram_loss_5: 0.17941467463970184
+INFO:__main__:vgg_loss_0: 11.9807767868042
+INFO:__main__:vgg_loss_1: 28.976285934448242
+INFO:__main__:vgg_loss_2: 40.867698669433594
+INFO:__main__:vgg_loss_3: 29.05699920654297
+INFO:__main__:vgg_loss_4: 51.03157424926758
+INFO:__main__:vgg_loss_5: 1.7021023035049438
+INFO:__main__:validation_loss: 1376.5712890625
+INFO:__main__:global_step: 35751
+INFO:__main__:kl_loss: 8451.173828125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1559.8070068359375
+INFO:__main__:loss: 1559.8154296875
+INFO:__main__:vgg_gram_loss_0: 6.388759136199951
+INFO:__main__:vgg_gram_loss_1: 16.789342880249023
+INFO:__main__:vgg_gram_loss_2: 19.687841415405273
+INFO:__main__:vgg_gram_loss_3: 16.294286727905273
+INFO:__main__:vgg_gram_loss_4: 76.1805648803711
+INFO:__main__:vgg_gram_loss_5: 0.17934273183345795
+INFO:__main__:vgg_loss_0: 13.143385887145996
+INFO:__main__:vgg_loss_1: 29.143680572509766
+INFO:__main__:vgg_loss_2: 42.55323791503906
+INFO:__main__:vgg_loss_3: 31.910526275634766
+INFO:__main__:vgg_loss_4: 57.76618576049805
+INFO:__main__:vgg_loss_5: 1.9242489337921143
+INFO:__main__:validation_loss: 2288.017578125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 36001
+INFO:__main__:kl_loss: 7553.9130859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1746.185302734375
+INFO:__main__:loss: 1746.19287109375
+INFO:__main__:vgg_gram_loss_0: 9.255935668945312
+INFO:__main__:vgg_gram_loss_1: 20.793149948120117
+INFO:__main__:vgg_gram_loss_2: 25.787322998046875
+INFO:__main__:vgg_gram_loss_3: 22.0164852142334
+INFO:__main__:vgg_gram_loss_4: 88.27578735351562
+INFO:__main__:vgg_gram_loss_5: 0.18374089896678925
+INFO:__main__:vgg_loss_0: 13.735885620117188
+INFO:__main__:vgg_loss_1: 34.52604293823242
+INFO:__main__:vgg_loss_2: 47.120784759521484
+INFO:__main__:vgg_loss_3: 31.969921112060547
+INFO:__main__:vgg_loss_4: 53.78569793701172
+INFO:__main__:vgg_loss_5: 1.7862669229507446
+INFO:__main__:validation_loss: 1900.9381103515625
+INFO:__main__:global_step: 36251
+INFO:__main__:kl_loss: 7772.583984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1503.9161376953125
+INFO:__main__:loss: 1503.9239501953125
+INFO:__main__:vgg_gram_loss_0: 4.804553031921387
+INFO:__main__:vgg_gram_loss_1: 14.753111839294434
+INFO:__main__:vgg_gram_loss_2: 21.612346649169922
+INFO:__main__:vgg_gram_loss_3: 16.24247932434082
+INFO:__main__:vgg_gram_loss_4: 74.70458221435547
+INFO:__main__:vgg_gram_loss_5: 0.17574135959148407
+INFO:__main__:vgg_loss_0: 12.151427268981934
+INFO:__main__:vgg_loss_1: 29.248201370239258
+INFO:__main__:vgg_loss_2: 41.967437744140625
+INFO:__main__:vgg_loss_3: 29.91663932800293
+INFO:__main__:vgg_loss_4: 53.397701263427734
+INFO:__main__:vgg_loss_5: 1.8090234994888306
+INFO:__main__:validation_loss: 1785.992431640625
+INFO:__main__:global_step: 36501
+INFO:__main__:kl_loss: 7209.14208984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1599.328369140625
+INFO:__main__:loss: 1599.3355712890625
+INFO:__main__:vgg_gram_loss_0: 7.356851100921631
+INFO:__main__:vgg_gram_loss_1: 20.298858642578125
+INFO:__main__:vgg_gram_loss_2: 21.875749588012695
+INFO:__main__:vgg_gram_loss_3: 17.733182907104492
+INFO:__main__:vgg_gram_loss_4: 80.12234497070312
+INFO:__main__:vgg_gram_loss_5: 0.19216053187847137
+INFO:__main__:vgg_loss_0: 12.841423034667969
+INFO:__main__:vgg_loss_1: 32.35765838623047
+INFO:__main__:vgg_loss_2: 42.91802978515625
+INFO:__main__:vgg_loss_3: 30.298646926879883
+INFO:__main__:vgg_loss_4: 52.12350082397461
+INFO:__main__:vgg_loss_5: 1.747230887413025
+INFO:__main__:validation_loss: 1815.7471923828125
+INFO:__main__:global_step: 36751
+INFO:__main__:kl_loss: 9152.4345703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1610.016845703125
+INFO:__main__:loss: 1610.0260009765625
+INFO:__main__:vgg_gram_loss_0: 7.259591579437256
+INFO:__main__:vgg_gram_loss_1: 26.751981735229492
+INFO:__main__:vgg_gram_loss_2: 24.15738868713379
+INFO:__main__:vgg_gram_loss_3: 19.41010093688965
+INFO:__main__:vgg_gram_loss_4: 76.37455749511719
+INFO:__main__:vgg_gram_loss_5: 0.15342143177986145
+INFO:__main__:vgg_loss_0: 13.114911079406738
+INFO:__main__:vgg_loss_1: 34.795658111572266
+INFO:__main__:vgg_loss_2: 42.658931732177734
+INFO:__main__:vgg_loss_3: 27.849403381347656
+INFO:__main__:vgg_loss_4: 47.87040710449219
+INFO:__main__:vgg_loss_5: 1.6070008277893066
+INFO:__main__:validation_loss: 1735.09716796875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 37001
+INFO:__main__:kl_loss: 7648.14599609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1302.353759765625
+INFO:__main__:loss: 1302.3614501953125
+INFO:__main__:vgg_gram_loss_0: 7.602331638336182
+INFO:__main__:vgg_gram_loss_1: 14.79955005645752
+INFO:__main__:vgg_gram_loss_2: 18.46120262145996
+INFO:__main__:vgg_gram_loss_3: 14.231940269470215
+INFO:__main__:vgg_gram_loss_4: 65.24879455566406
+INFO:__main__:vgg_gram_loss_5: 0.14482997357845306
+INFO:__main__:vgg_loss_0: 9.934188842773438
+INFO:__main__:vgg_loss_1: 23.894054412841797
+INFO:__main__:vgg_loss_2: 34.83245849609375
+INFO:__main__:vgg_loss_3: 25.21493911743164
+INFO:__main__:vgg_loss_4: 44.64652633666992
+INFO:__main__:vgg_loss_5: 1.4599418640136719
+INFO:__main__:validation_loss: 1903.701416015625
+INFO:__main__:global_step: 37251
+INFO:__main__:kl_loss: 9111.056640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1678.5345458984375
+INFO:__main__:loss: 1678.543701171875
+INFO:__main__:vgg_gram_loss_0: 6.289675235748291
+INFO:__main__:vgg_gram_loss_1: 29.444677352905273
+INFO:__main__:vgg_gram_loss_2: 33.45845031738281
+INFO:__main__:vgg_gram_loss_3: 23.495359420776367
+INFO:__main__:vgg_gram_loss_4: 77.74526977539062
+INFO:__main__:vgg_gram_loss_5: 0.16624127328395844
+INFO:__main__:vgg_loss_0: 11.726822853088379
+INFO:__main__:vgg_loss_1: 29.57559585571289
+INFO:__main__:vgg_loss_2: 41.910484313964844
+INFO:__main__:vgg_loss_3: 29.389162063598633
+INFO:__main__:vgg_loss_4: 50.77119064331055
+INFO:__main__:vgg_loss_5: 1.7339656352996826
+INFO:__main__:validation_loss: 2013.730224609375
+INFO:__main__:global_step: 37501
+INFO:__main__:kl_loss: 10201.34765625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1423.797119140625
+INFO:__main__:loss: 1423.807373046875
+INFO:__main__:vgg_gram_loss_0: 7.088632106781006
+INFO:__main__:vgg_gram_loss_1: 18.344762802124023
+INFO:__main__:vgg_gram_loss_2: 19.61504554748535
+INFO:__main__:vgg_gram_loss_3: 15.51315975189209
+INFO:__main__:vgg_gram_loss_4: 72.98799896240234
+INFO:__main__:vgg_gram_loss_5: 0.1583849936723709
+INFO:__main__:vgg_loss_0: 10.556436538696289
+INFO:__main__:vgg_loss_1: 26.214061737060547
+INFO:__main__:vgg_loss_2: 38.40428924560547
+INFO:__main__:vgg_loss_3: 27.458547592163086
+INFO:__main__:vgg_loss_4: 46.876461029052734
+INFO:__main__:vgg_loss_5: 1.5416526794433594
+INFO:__main__:validation_loss: 1880.207275390625
+INFO:__main__:global_step: 37751
+INFO:__main__:kl_loss: 8125.533203125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1850.907958984375
+INFO:__main__:loss: 1850.9161376953125
+INFO:__main__:vgg_gram_loss_0: 11.285836219787598
+INFO:__main__:vgg_gram_loss_1: 30.047155380249023
+INFO:__main__:vgg_gram_loss_2: 29.37656021118164
+INFO:__main__:vgg_gram_loss_3: 22.120805740356445
+INFO:__main__:vgg_gram_loss_4: 91.99359130859375
+INFO:__main__:vgg_gram_loss_5: 0.1854284703731537
+INFO:__main__:vgg_loss_0: 13.541316986083984
+INFO:__main__:vgg_loss_1: 31.008007049560547
+INFO:__main__:vgg_loss_2: 45.592323303222656
+INFO:__main__:vgg_loss_3: 33.829322814941406
+INFO:__main__:vgg_loss_4: 59.274654388427734
+INFO:__main__:vgg_loss_5: 1.9265990257263184
+INFO:__main__:validation_loss: 1972.7264404296875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 38001
+INFO:__main__:kl_loss: 9419.595703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1733.5814208984375
+INFO:__main__:loss: 1733.5908203125
+INFO:__main__:vgg_gram_loss_0: 12.361004829406738
+INFO:__main__:vgg_gram_loss_1: 23.803571701049805
+INFO:__main__:vgg_gram_loss_2: 23.572237014770508
+INFO:__main__:vgg_gram_loss_3: 18.59601402282715
+INFO:__main__:vgg_gram_loss_4: 86.97547912597656
+INFO:__main__:vgg_gram_loss_5: 0.19118832051753998
+INFO:__main__:vgg_loss_0: 12.74040412902832
+INFO:__main__:vgg_loss_1: 30.129796981811523
+INFO:__main__:vgg_loss_2: 45.47703170776367
+INFO:__main__:vgg_loss_3: 32.6150016784668
+INFO:__main__:vgg_loss_4: 58.343318939208984
+INFO:__main__:vgg_loss_5: 1.9112049341201782
+INFO:__main__:validation_loss: 2338.38916015625
+INFO:__main__:global_step: 38251
+INFO:__main__:kl_loss: 9518.783203125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 2060.427490234375
+INFO:__main__:loss: 2060.43701171875
+INFO:__main__:vgg_gram_loss_0: 10.313047409057617
+INFO:__main__:vgg_gram_loss_1: 27.976511001586914
+INFO:__main__:vgg_gram_loss_2: 38.32465362548828
+INFO:__main__:vgg_gram_loss_3: 27.022119522094727
+INFO:__main__:vgg_gram_loss_4: 97.69241333007812
+INFO:__main__:vgg_gram_loss_5: 0.2003655731678009
+INFO:__main__:vgg_loss_0: 15.831891059875488
+INFO:__main__:vgg_loss_1: 39.451725006103516
+INFO:__main__:vgg_loss_2: 54.24762725830078
+INFO:__main__:vgg_loss_3: 36.66573715209961
+INFO:__main__:vgg_loss_4: 62.35550308227539
+INFO:__main__:vgg_loss_5: 2.00393009185791
+INFO:__main__:validation_loss: 2011.1722412109375
+INFO:__main__:global_step: 38501
+INFO:__main__:kl_loss: 9474.318359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1508.028564453125
+INFO:__main__:loss: 1508.0380859375
+INFO:__main__:vgg_gram_loss_0: 7.7902727127075195
+INFO:__main__:vgg_gram_loss_1: 22.38382911682129
+INFO:__main__:vgg_gram_loss_2: 20.956754684448242
+INFO:__main__:vgg_gram_loss_3: 19.129777908325195
+INFO:__main__:vgg_gram_loss_4: 75.7132797241211
+INFO:__main__:vgg_gram_loss_5: 0.17522607743740082
+INFO:__main__:vgg_loss_0: 10.89272403717041
+INFO:__main__:vgg_loss_1: 26.571640014648438
+INFO:__main__:vgg_loss_2: 38.16643524169922
+INFO:__main__:vgg_loss_3: 28.097518920898438
+INFO:__main__:vgg_loss_4: 50.052101135253906
+INFO:__main__:vgg_loss_5: 1.6761561632156372
+INFO:__main__:validation_loss: 1843.180419921875
+INFO:__main__:global_step: 38751
+INFO:__main__:kl_loss: 9249.4697265625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1742.8668212890625
+INFO:__main__:loss: 1742.8760986328125
+INFO:__main__:vgg_gram_loss_0: 7.83588171005249
+INFO:__main__:vgg_gram_loss_1: 20.856687545776367
+INFO:__main__:vgg_gram_loss_2: 33.193946838378906
+INFO:__main__:vgg_gram_loss_3: 26.880416870117188
+INFO:__main__:vgg_gram_loss_4: 87.89495086669922
+INFO:__main__:vgg_gram_loss_5: 0.19870135188102722
+INFO:__main__:vgg_loss_0: 11.750406265258789
+INFO:__main__:vgg_loss_1: 29.24363136291504
+INFO:__main__:vgg_loss_2: 43.41839599609375
+INFO:__main__:vgg_loss_3: 31.520845413208008
+INFO:__main__:vgg_loss_4: 53.951114654541016
+INFO:__main__:vgg_loss_5: 1.8283920288085938
+INFO:__main__:validation_loss: 2239.486083984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 39001
+INFO:__main__:kl_loss: 7569.99658203125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1423.7969970703125
+INFO:__main__:loss: 1423.8045654296875
+INFO:__main__:vgg_gram_loss_0: 9.573261260986328
+INFO:__main__:vgg_gram_loss_1: 16.354341506958008
+INFO:__main__:vgg_gram_loss_2: 18.250646591186523
+INFO:__main__:vgg_gram_loss_3: 15.152325630187988
+INFO:__main__:vgg_gram_loss_4: 72.42867279052734
+INFO:__main__:vgg_gram_loss_5: 0.17282617092132568
+INFO:__main__:vgg_loss_0: 10.599743843078613
+INFO:__main__:vgg_loss_1: 25.255971908569336
+INFO:__main__:vgg_loss_2: 37.571231842041016
+INFO:__main__:vgg_loss_3: 28.005123138427734
+INFO:__main__:vgg_loss_4: 49.75764083862305
+INFO:__main__:vgg_loss_5: 1.6376086473464966
+INFO:__main__:validation_loss: 1635.4595947265625
+INFO:__main__:global_step: 39251
+INFO:__main__:kl_loss: 8055.4599609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1606.0399169921875
+INFO:__main__:loss: 1606.0479736328125
+INFO:__main__:vgg_gram_loss_0: 10.887603759765625
+INFO:__main__:vgg_gram_loss_1: 20.1430606842041
+INFO:__main__:vgg_gram_loss_2: 25.136465072631836
+INFO:__main__:vgg_gram_loss_3: 19.698516845703125
+INFO:__main__:vgg_gram_loss_4: 80.75541687011719
+INFO:__main__:vgg_gram_loss_5: 0.16688372194766998
+INFO:__main__:vgg_loss_0: 11.602705955505371
+INFO:__main__:vgg_loss_1: 28.676408767700195
+INFO:__main__:vgg_loss_2: 41.04060363769531
+INFO:__main__:vgg_loss_3: 29.47882652282715
+INFO:__main__:vgg_loss_4: 51.91919708251953
+INFO:__main__:vgg_loss_5: 1.7022839784622192
+INFO:__main__:validation_loss: 1720.8876953125
+INFO:__main__:global_step: 39501
+INFO:__main__:kl_loss: 9407.181640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1675.4443359375
+INFO:__main__:loss: 1675.4537353515625
+INFO:__main__:vgg_gram_loss_0: 6.195657253265381
+INFO:__main__:vgg_gram_loss_1: 20.00478172302246
+INFO:__main__:vgg_gram_loss_2: 33.11785888671875
+INFO:__main__:vgg_gram_loss_3: 24.370134353637695
+INFO:__main__:vgg_gram_loss_4: 81.1272964477539
+INFO:__main__:vgg_gram_loss_5: 0.1531304270029068
+INFO:__main__:vgg_loss_0: 12.192713737487793
+INFO:__main__:vgg_loss_1: 31.965723037719727
+INFO:__main__:vgg_loss_2: 45.188446044921875
+INFO:__main__:vgg_loss_3: 30.495031356811523
+INFO:__main__:vgg_loss_4: 48.73186111450195
+INFO:__main__:vgg_loss_5: 1.5462100505828857
+INFO:__main__:validation_loss: 1805.4560546875
+INFO:__main__:global_step: 39751
+INFO:__main__:kl_loss: 8921.2373046875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1445.104736328125
+INFO:__main__:loss: 1445.1136474609375
+INFO:__main__:vgg_gram_loss_0: 6.228761196136475
+INFO:__main__:vgg_gram_loss_1: 18.240026473999023
+INFO:__main__:vgg_gram_loss_2: 21.422693252563477
+INFO:__main__:vgg_gram_loss_3: 15.411415100097656
+INFO:__main__:vgg_gram_loss_4: 68.06610107421875
+INFO:__main__:vgg_gram_loss_5: 0.16765783727169037
+INFO:__main__:vgg_loss_0: 11.41713809967041
+INFO:__main__:vgg_loss_1: 27.90165901184082
+INFO:__main__:vgg_loss_2: 40.970943450927734
+INFO:__main__:vgg_loss_3: 28.40261459350586
+INFO:__main__:vgg_loss_4: 49.144073486328125
+INFO:__main__:vgg_loss_5: 1.6478737592697144
+INFO:__main__:validation_loss: 2044.31787109375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 40001
+INFO:__main__:kl_loss: 8835.5859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1756.416015625
+INFO:__main__:loss: 1756.4248046875
+INFO:__main__:vgg_gram_loss_0: 13.285600662231445
+INFO:__main__:vgg_gram_loss_1: 22.949234008789062
+INFO:__main__:vgg_gram_loss_2: 26.83192253112793
+INFO:__main__:vgg_gram_loss_3: 21.025815963745117
+INFO:__main__:vgg_gram_loss_4: 88.21632385253906
+INFO:__main__:vgg_gram_loss_5: 0.18179000914096832
+INFO:__main__:vgg_loss_0: 13.19233226776123
+INFO:__main__:vgg_loss_1: 31.744434356689453
+INFO:__main__:vgg_loss_2: 44.59366989135742
+INFO:__main__:vgg_loss_3: 31.583251953125
+INFO:__main__:vgg_loss_4: 55.838191986083984
+INFO:__main__:vgg_loss_5: 1.8406068086624146
+INFO:__main__:validation_loss: 1870.2249755859375
+INFO:__main__:global_step: 40251
+INFO:__main__:kl_loss: 9479.92578125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1748.8116455078125
+INFO:__main__:loss: 1748.8211669921875
+INFO:__main__:vgg_gram_loss_0: 7.578904151916504
+INFO:__main__:vgg_gram_loss_1: 19.574827194213867
+INFO:__main__:vgg_gram_loss_2: 28.963642120361328
+INFO:__main__:vgg_gram_loss_3: 22.991683959960938
+INFO:__main__:vgg_gram_loss_4: 85.82445526123047
+INFO:__main__:vgg_gram_loss_5: 0.17482519149780273
+INFO:__main__:vgg_loss_0: 13.565220832824707
+INFO:__main__:vgg_loss_1: 33.028968811035156
+INFO:__main__:vgg_loss_2: 46.2962532043457
+INFO:__main__:vgg_loss_3: 32.74042510986328
+INFO:__main__:vgg_loss_4: 57.112606048583984
+INFO:__main__:vgg_loss_5: 1.9105504751205444
+INFO:__main__:validation_loss: 1610.255859375
+INFO:__main__:global_step: 40501
+INFO:__main__:kl_loss: 8865.5
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1214.0989990234375
+INFO:__main__:loss: 1214.10791015625
+INFO:__main__:vgg_gram_loss_0: 5.549455165863037
+INFO:__main__:vgg_gram_loss_1: 11.609081268310547
+INFO:__main__:vgg_gram_loss_2: 16.091876983642578
+INFO:__main__:vgg_gram_loss_3: 12.829075813293457
+INFO:__main__:vgg_gram_loss_4: 61.15956497192383
+INFO:__main__:vgg_gram_loss_5: 0.14389389753341675
+INFO:__main__:vgg_loss_0: 8.696565628051758
+INFO:__main__:vgg_loss_1: 21.34469223022461
+INFO:__main__:vgg_loss_2: 33.75932312011719
+INFO:__main__:vgg_loss_3: 25.207828521728516
+INFO:__main__:vgg_loss_4: 44.95752716064453
+INFO:__main__:vgg_loss_5: 1.4708985090255737
+INFO:__main__:validation_loss: 1877.5142822265625
+INFO:__main__:global_step: 40751
+INFO:__main__:kl_loss: 10130.162109375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1719.4500732421875
+INFO:__main__:loss: 1719.460205078125
+INFO:__main__:vgg_gram_loss_0: 10.33341121673584
+INFO:__main__:vgg_gram_loss_1: 21.72492218017578
+INFO:__main__:vgg_gram_loss_2: 26.83856773376465
+INFO:__main__:vgg_gram_loss_3: 21.325246810913086
+INFO:__main__:vgg_gram_loss_4: 87.40166473388672
+INFO:__main__:vgg_gram_loss_5: 0.1841447800397873
+INFO:__main__:vgg_loss_0: 12.617207527160645
+INFO:__main__:vgg_loss_1: 31.58260726928711
+INFO:__main__:vgg_loss_2: 44.888519287109375
+INFO:__main__:vgg_loss_3: 31.738065719604492
+INFO:__main__:vgg_loss_4: 53.523948669433594
+INFO:__main__:vgg_loss_5: 1.7316975593566895
+INFO:__main__:validation_loss: 1668.9146728515625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 41001
+INFO:__main__:kl_loss: 8297.326171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1756.654052734375
+INFO:__main__:loss: 1756.662353515625
+INFO:__main__:vgg_gram_loss_0: 10.775683403015137
+INFO:__main__:vgg_gram_loss_1: 23.95917320251465
+INFO:__main__:vgg_gram_loss_2: 25.010766983032227
+INFO:__main__:vgg_gram_loss_3: 21.397451400756836
+INFO:__main__:vgg_gram_loss_4: 86.80598449707031
+INFO:__main__:vgg_gram_loss_5: 0.18591319024562836
+INFO:__main__:vgg_loss_0: 13.500639915466309
+INFO:__main__:vgg_loss_1: 33.56721496582031
+INFO:__main__:vgg_loss_2: 46.034820556640625
+INFO:__main__:vgg_loss_3: 32.26215362548828
+INFO:__main__:vgg_loss_4: 55.98773956298828
+INFO:__main__:vgg_loss_5: 1.843262791633606
+INFO:__main__:validation_loss: 1792.62744140625
+INFO:__main__:global_step: 41251
+INFO:__main__:kl_loss: 9321.31640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1622.11962890625
+INFO:__main__:loss: 1622.12890625
+INFO:__main__:vgg_gram_loss_0: 5.1870293617248535
+INFO:__main__:vgg_gram_loss_1: 17.641048431396484
+INFO:__main__:vgg_gram_loss_2: 24.76276397705078
+INFO:__main__:vgg_gram_loss_3: 21.049531936645508
+INFO:__main__:vgg_gram_loss_4: 85.08516693115234
+INFO:__main__:vgg_gram_loss_5: 0.17605161666870117
+INFO:__main__:vgg_loss_0: 12.118850708007812
+INFO:__main__:vgg_loss_1: 30.45805549621582
+INFO:__main__:vgg_loss_2: 43.062740325927734
+INFO:__main__:vgg_loss_3: 30.306379318237305
+INFO:__main__:vgg_loss_4: 52.844722747802734
+INFO:__main__:vgg_loss_5: 1.731568694114685
+INFO:__main__:validation_loss: 2122.1435546875
+INFO:__main__:global_step: 41501
+INFO:__main__:kl_loss: 9629.1669921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1729.3316650390625
+INFO:__main__:loss: 1729.34130859375
+INFO:__main__:vgg_gram_loss_0: 10.370546340942383
+INFO:__main__:vgg_gram_loss_1: 22.965070724487305
+INFO:__main__:vgg_gram_loss_2: 32.55009841918945
+INFO:__main__:vgg_gram_loss_3: 23.038524627685547
+INFO:__main__:vgg_gram_loss_4: 82.88329315185547
+INFO:__main__:vgg_gram_loss_5: 0.162612184882164
+INFO:__main__:vgg_loss_0: 13.443943977355957
+INFO:__main__:vgg_loss_1: 31.29157829284668
+INFO:__main__:vgg_loss_2: 43.43289566040039
+INFO:__main__:vgg_loss_3: 31.020904541015625
+INFO:__main__:vgg_loss_4: 52.96934509277344
+INFO:__main__:vgg_loss_5: 1.7375121116638184
+INFO:__main__:validation_loss: 1720.43505859375
+INFO:__main__:global_step: 41751
+INFO:__main__:kl_loss: 7760.45458984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1171.41845703125
+INFO:__main__:loss: 1171.42626953125
+INFO:__main__:vgg_gram_loss_0: 5.524149417877197
+INFO:__main__:vgg_gram_loss_1: 14.368407249450684
+INFO:__main__:vgg_gram_loss_2: 14.979059219360352
+INFO:__main__:vgg_gram_loss_3: 12.026633262634277
+INFO:__main__:vgg_gram_loss_4: 60.20589828491211
+INFO:__main__:vgg_gram_loss_5: 0.13775920867919922
+INFO:__main__:vgg_loss_0: 8.54099178314209
+INFO:__main__:vgg_loss_1: 20.751142501831055
+INFO:__main__:vgg_loss_2: 31.884323120117188
+INFO:__main__:vgg_loss_3: 23.20992660522461
+INFO:__main__:vgg_loss_4: 41.266387939453125
+INFO:__main__:vgg_loss_5: 1.3890204429626465
+INFO:__main__:validation_loss: 2049.815185546875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 42001
+INFO:__main__:kl_loss: 10774.4921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1554.42041015625
+INFO:__main__:loss: 1554.43115234375
+INFO:__main__:vgg_gram_loss_0: 10.117583274841309
+INFO:__main__:vgg_gram_loss_1: 22.148685455322266
+INFO:__main__:vgg_gram_loss_2: 21.80250358581543
+INFO:__main__:vgg_gram_loss_3: 17.378145217895508
+INFO:__main__:vgg_gram_loss_4: 79.85076141357422
+INFO:__main__:vgg_gram_loss_5: 0.18269981443881989
+INFO:__main__:vgg_loss_0: 11.245356559753418
+INFO:__main__:vgg_loss_1: 26.620996475219727
+INFO:__main__:vgg_loss_2: 38.28331756591797
+INFO:__main__:vgg_loss_3: 29.019271850585938
+INFO:__main__:vgg_loss_4: 52.459503173828125
+INFO:__main__:vgg_loss_5: 1.7752699851989746
+INFO:__main__:validation_loss: 2204.73046875
+INFO:__main__:global_step: 42251
+INFO:__main__:kl_loss: 9690.2119140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1595.619140625
+INFO:__main__:loss: 1595.6287841796875
+INFO:__main__:vgg_gram_loss_0: 6.687384128570557
+INFO:__main__:vgg_gram_loss_1: 16.456064224243164
+INFO:__main__:vgg_gram_loss_2: 20.837387084960938
+INFO:__main__:vgg_gram_loss_3: 18.50313949584961
+INFO:__main__:vgg_gram_loss_4: 84.68807220458984
+INFO:__main__:vgg_gram_loss_5: 0.18541781604290009
+INFO:__main__:vgg_loss_0: 11.752615928649902
+INFO:__main__:vgg_loss_1: 27.975170135498047
+INFO:__main__:vgg_loss_2: 42.3662223815918
+INFO:__main__:vgg_loss_3: 31.381454467773438
+INFO:__main__:vgg_loss_4: 56.4162483215332
+INFO:__main__:vgg_loss_5: 1.8746510744094849
+INFO:__main__:validation_loss: 2143.83447265625
+INFO:__main__:global_step: 42501
+INFO:__main__:kl_loss: 9934.81640625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1872.5047607421875
+INFO:__main__:loss: 1872.5146484375
+INFO:__main__:vgg_gram_loss_0: 12.773937225341797
+INFO:__main__:vgg_gram_loss_1: 30.18731689453125
+INFO:__main__:vgg_gram_loss_2: 29.598819732666016
+INFO:__main__:vgg_gram_loss_3: 25.712451934814453
+INFO:__main__:vgg_gram_loss_4: 96.12899780273438
+INFO:__main__:vgg_gram_loss_5: 0.1937173455953598
+INFO:__main__:vgg_loss_0: 13.41330623626709
+INFO:__main__:vgg_loss_1: 32.328792572021484
+INFO:__main__:vgg_loss_2: 44.687564849853516
+INFO:__main__:vgg_loss_3: 32.12128829956055
+INFO:__main__:vgg_loss_4: 55.499454498291016
+INFO:__main__:vgg_loss_5: 1.8553204536437988
+INFO:__main__:validation_loss: 2310.297607421875
+INFO:__main__:global_step: 42751
+INFO:__main__:kl_loss: 9581.1416015625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1596.0203857421875
+INFO:__main__:loss: 1596.0299072265625
+INFO:__main__:vgg_gram_loss_0: 5.961778163909912
+INFO:__main__:vgg_gram_loss_1: 25.3477725982666
+INFO:__main__:vgg_gram_loss_2: 22.447004318237305
+INFO:__main__:vgg_gram_loss_3: 17.86504554748535
+INFO:__main__:vgg_gram_loss_4: 79.71936798095703
+INFO:__main__:vgg_gram_loss_5: 0.17816276848316193
+INFO:__main__:vgg_loss_0: 13.08590316772461
+INFO:__main__:vgg_loss_1: 32.5493049621582
+INFO:__main__:vgg_loss_2: 41.170955657958984
+INFO:__main__:vgg_loss_3: 29.197307586669922
+INFO:__main__:vgg_loss_4: 50.03356170654297
+INFO:__main__:vgg_loss_5: 1.647894263267517
+INFO:__main__:validation_loss: 1669.0289306640625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 43001
+INFO:__main__:kl_loss: 8766.6083984375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1422.8505859375
+INFO:__main__:loss: 1422.859375
+INFO:__main__:vgg_gram_loss_0: 13.175976753234863
+INFO:__main__:vgg_gram_loss_1: 20.616771697998047
+INFO:__main__:vgg_gram_loss_2: 19.96365737915039
+INFO:__main__:vgg_gram_loss_3: 15.677929878234863
+INFO:__main__:vgg_gram_loss_4: 68.13050079345703
+INFO:__main__:vgg_gram_loss_5: 0.1573638916015625
+INFO:__main__:vgg_loss_0: 10.43492603302002
+INFO:__main__:vgg_loss_1: 24.882993698120117
+INFO:__main__:vgg_loss_2: 36.53907775878906
+INFO:__main__:vgg_loss_3: 26.51683235168457
+INFO:__main__:vgg_loss_4: 46.914371490478516
+INFO:__main__:vgg_loss_5: 1.5597275495529175
+INFO:__main__:validation_loss: 2067.302001953125
+INFO:__main__:global_step: 43251
+INFO:__main__:kl_loss: 9174.779296875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1414.285400390625
+INFO:__main__:loss: 1414.2945556640625
+INFO:__main__:vgg_gram_loss_0: 8.293191909790039
+INFO:__main__:vgg_gram_loss_1: 19.458078384399414
+INFO:__main__:vgg_gram_loss_2: 25.102001190185547
+INFO:__main__:vgg_gram_loss_3: 17.223682403564453
+INFO:__main__:vgg_gram_loss_4: 66.61102294921875
+INFO:__main__:vgg_gram_loss_5: 0.14332716166973114
+INFO:__main__:vgg_loss_0: 10.430252075195312
+INFO:__main__:vgg_loss_1: 26.313987731933594
+INFO:__main__:vgg_loss_2: 36.42021179199219
+INFO:__main__:vgg_loss_3: 25.402538299560547
+INFO:__main__:vgg_loss_4: 45.8823127746582
+INFO:__main__:vgg_loss_5: 1.5764851570129395
+INFO:__main__:validation_loss: 1737.7913818359375
+INFO:__main__:global_step: 43501
+INFO:__main__:kl_loss: 9088.90625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1287.9010009765625
+INFO:__main__:loss: 1287.9100341796875
+INFO:__main__:vgg_gram_loss_0: 6.474820613861084
+INFO:__main__:vgg_gram_loss_1: 14.208205223083496
+INFO:__main__:vgg_gram_loss_2: 18.201261520385742
+INFO:__main__:vgg_gram_loss_3: 14.232101440429688
+INFO:__main__:vgg_gram_loss_4: 65.35556030273438
+INFO:__main__:vgg_gram_loss_5: 0.14019489288330078
+INFO:__main__:vgg_loss_0: 9.71634578704834
+INFO:__main__:vgg_loss_1: 22.936613082885742
+INFO:__main__:vgg_loss_2: 34.41395950317383
+INFO:__main__:vgg_loss_3: 25.192184448242188
+INFO:__main__:vgg_loss_4: 45.206905364990234
+INFO:__main__:vgg_loss_5: 1.5020639896392822
+INFO:__main__:validation_loss: 2021.074462890625
+INFO:__main__:global_step: 43751
+INFO:__main__:kl_loss: 9391.5068359375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1660.9571533203125
+INFO:__main__:loss: 1660.966552734375
+INFO:__main__:vgg_gram_loss_0: 9.486883163452148
+INFO:__main__:vgg_gram_loss_1: 21.380613327026367
+INFO:__main__:vgg_gram_loss_2: 24.187665939331055
+INFO:__main__:vgg_gram_loss_3: 19.412879943847656
+INFO:__main__:vgg_gram_loss_4: 81.29649353027344
+INFO:__main__:vgg_gram_loss_5: 0.18476049602031708
+INFO:__main__:vgg_loss_0: 12.340121269226074
+INFO:__main__:vgg_loss_1: 31.325048446655273
+INFO:__main__:vgg_loss_2: 45.184329986572266
+INFO:__main__:vgg_loss_3: 31.754453659057617
+INFO:__main__:vgg_loss_4: 53.9002799987793
+INFO:__main__:vgg_loss_5: 1.7379114627838135
+INFO:__main__:validation_loss: 1886.210693359375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 44001
+INFO:__main__:kl_loss: 9578.005859375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1441.51416015625
+INFO:__main__:loss: 1441.523681640625
+INFO:__main__:vgg_gram_loss_0: 8.3568754196167
+INFO:__main__:vgg_gram_loss_1: 18.725130081176758
+INFO:__main__:vgg_gram_loss_2: 18.568134307861328
+INFO:__main__:vgg_gram_loss_3: 15.511079788208008
+INFO:__main__:vgg_gram_loss_4: 72.5020751953125
+INFO:__main__:vgg_gram_loss_5: 0.15925979614257812
+INFO:__main__:vgg_loss_0: 11.454483032226562
+INFO:__main__:vgg_loss_1: 28.54051971435547
+INFO:__main__:vgg_loss_2: 38.66178512573242
+INFO:__main__:vgg_loss_3: 27.00679588317871
+INFO:__main__:vgg_loss_4: 47.23535919189453
+INFO:__main__:vgg_loss_5: 1.5813331604003906
+INFO:__main__:validation_loss: 1775.327880859375
+INFO:__main__:global_step: 44251
+INFO:__main__:kl_loss: 8632.037109375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1457.911376953125
+INFO:__main__:loss: 1457.9200439453125
+INFO:__main__:vgg_gram_loss_0: 7.811997413635254
+INFO:__main__:vgg_gram_loss_1: 17.430986404418945
+INFO:__main__:vgg_gram_loss_2: 19.178979873657227
+INFO:__main__:vgg_gram_loss_3: 15.022111892700195
+INFO:__main__:vgg_gram_loss_4: 71.64246368408203
+INFO:__main__:vgg_gram_loss_5: 0.16500459611415863
+INFO:__main__:vgg_loss_0: 10.91763973236084
+INFO:__main__:vgg_loss_1: 27.38864517211914
+INFO:__main__:vgg_loss_2: 39.3030891418457
+INFO:__main__:vgg_loss_3: 28.41068458557129
+INFO:__main__:vgg_loss_4: 52.53445053100586
+INFO:__main__:vgg_loss_5: 1.776222825050354
+INFO:__main__:validation_loss: 1784.4569091796875
+INFO:__main__:global_step: 44501
+INFO:__main__:kl_loss: 8698.96484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1622.139892578125
+INFO:__main__:loss: 1622.1485595703125
+INFO:__main__:vgg_gram_loss_0: 10.778220176696777
+INFO:__main__:vgg_gram_loss_1: 23.28611183166504
+INFO:__main__:vgg_gram_loss_2: 24.35749053955078
+INFO:__main__:vgg_gram_loss_3: 18.789602279663086
+INFO:__main__:vgg_gram_loss_4: 80.2012710571289
+INFO:__main__:vgg_gram_loss_5: 0.17325326800346375
+INFO:__main__:vgg_loss_0: 12.196002960205078
+INFO:__main__:vgg_loss_1: 27.754016876220703
+INFO:__main__:vgg_loss_2: 40.98173141479492
+INFO:__main__:vgg_loss_3: 30.518558502197266
+INFO:__main__:vgg_loss_4: 53.61234664916992
+INFO:__main__:vgg_loss_5: 1.779394268989563
+INFO:__main__:validation_loss: 1742.3265380859375
+INFO:__main__:global_step: 44751
+INFO:__main__:kl_loss: 9614.53515625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1300.4881591796875
+INFO:__main__:loss: 1300.497802734375
+INFO:__main__:vgg_gram_loss_0: 6.188023090362549
+INFO:__main__:vgg_gram_loss_1: 15.605554580688477
+INFO:__main__:vgg_gram_loss_2: 16.89484214782715
+INFO:__main__:vgg_gram_loss_3: 14.300389289855957
+INFO:__main__:vgg_gram_loss_4: 65.00393676757812
+INFO:__main__:vgg_gram_loss_5: 0.14798857271671295
+INFO:__main__:vgg_loss_0: 9.03235149383545
+INFO:__main__:vgg_loss_1: 22.887588500976562
+INFO:__main__:vgg_loss_2: 35.74382400512695
+INFO:__main__:vgg_loss_3: 26.235004425048828
+INFO:__main__:vgg_loss_4: 46.511295318603516
+INFO:__main__:vgg_loss_5: 1.546831727027893
+INFO:__main__:validation_loss: 2100.8994140625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 45001
+INFO:__main__:kl_loss: 11119.951171875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1636.18994140625
+INFO:__main__:loss: 1636.2010498046875
+INFO:__main__:vgg_gram_loss_0: 6.257763862609863
+INFO:__main__:vgg_gram_loss_1: 21.747140884399414
+INFO:__main__:vgg_gram_loss_2: 27.795560836791992
+INFO:__main__:vgg_gram_loss_3: 21.7324275970459
+INFO:__main__:vgg_gram_loss_4: 79.32820129394531
+INFO:__main__:vgg_gram_loss_5: 0.17926429212093353
+INFO:__main__:vgg_loss_0: 12.734206199645996
+INFO:__main__:vgg_loss_1: 32.0416259765625
+INFO:__main__:vgg_loss_2: 42.97636795043945
+INFO:__main__:vgg_loss_3: 29.90683364868164
+INFO:__main__:vgg_loss_4: 50.79035186767578
+INFO:__main__:vgg_loss_5: 1.7482315301895142
+INFO:__main__:validation_loss: 2227.136962890625
+INFO:__main__:global_step: 45251
+INFO:__main__:kl_loss: 9197.736328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1275.852294921875
+INFO:__main__:loss: 1275.8614501953125
+INFO:__main__:vgg_gram_loss_0: 8.55854606628418
+INFO:__main__:vgg_gram_loss_1: 12.4595947265625
+INFO:__main__:vgg_gram_loss_2: 15.525347709655762
+INFO:__main__:vgg_gram_loss_3: 12.588106155395508
+INFO:__main__:vgg_gram_loss_4: 65.49786376953125
+INFO:__main__:vgg_gram_loss_5: 0.15774497389793396
+INFO:__main__:vgg_loss_0: 10.166051864624023
+INFO:__main__:vgg_loss_1: 23.966094970703125
+INFO:__main__:vgg_loss_2: 34.28947830200195
+INFO:__main__:vgg_loss_3: 24.786027908325195
+INFO:__main__:vgg_loss_4: 45.53057098388672
+INFO:__main__:vgg_loss_5: 1.6450568437576294
+INFO:__main__:validation_loss: 1712.7000732421875
+INFO:__main__:global_step: 45501
+INFO:__main__:kl_loss: 10470.8828125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1424.3133544921875
+INFO:__main__:loss: 1424.3238525390625
+INFO:__main__:vgg_gram_loss_0: 5.505692958831787
+INFO:__main__:vgg_gram_loss_1: 17.72892189025879
+INFO:__main__:vgg_gram_loss_2: 20.36162757873535
+INFO:__main__:vgg_gram_loss_3: 15.450578689575195
+INFO:__main__:vgg_gram_loss_4: 68.43685150146484
+INFO:__main__:vgg_gram_loss_5: 0.1570785492658615
+INFO:__main__:vgg_loss_0: 11.841625213623047
+INFO:__main__:vgg_loss_1: 27.814090728759766
+INFO:__main__:vgg_loss_2: 39.09339904785156
+INFO:__main__:vgg_loss_3: 27.879079818725586
+INFO:__main__:vgg_loss_4: 48.98435974121094
+INFO:__main__:vgg_loss_5: 1.6093655824661255
+INFO:__main__:validation_loss: 1818.6302490234375
+INFO:__main__:global_step: 45751
+INFO:__main__:kl_loss: 12194.9619140625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1556.3525390625
+INFO:__main__:loss: 1556.36474609375
+INFO:__main__:vgg_gram_loss_0: 10.379335403442383
+INFO:__main__:vgg_gram_loss_1: 21.33768653869629
+INFO:__main__:vgg_gram_loss_2: 22.973947525024414
+INFO:__main__:vgg_gram_loss_3: 18.441633224487305
+INFO:__main__:vgg_gram_loss_4: 76.26399230957031
+INFO:__main__:vgg_gram_loss_5: 0.17760609090328217
+INFO:__main__:vgg_loss_0: 11.58940601348877
+INFO:__main__:vgg_loss_1: 27.034364700317383
+INFO:__main__:vgg_loss_2: 39.85527801513672
+INFO:__main__:vgg_loss_3: 29.524471282958984
+INFO:__main__:vgg_loss_4: 51.941917419433594
+INFO:__main__:vgg_loss_5: 1.7508496046066284
+INFO:__main__:validation_loss: 1767.68408203125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 46001
+INFO:__main__:kl_loss: 12475.865234375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1439.7618408203125
+INFO:__main__:loss: 1439.7742919921875
+INFO:__main__:vgg_gram_loss_0: 8.620463371276855
+INFO:__main__:vgg_gram_loss_1: 21.148704528808594
+INFO:__main__:vgg_gram_loss_2: 20.159622192382812
+INFO:__main__:vgg_gram_loss_3: 16.194292068481445
+INFO:__main__:vgg_gram_loss_4: 75.39995574951172
+INFO:__main__:vgg_gram_loss_5: 0.16231608390808105
+INFO:__main__:vgg_loss_0: 10.062175750732422
+INFO:__main__:vgg_loss_1: 24.90937614440918
+INFO:__main__:vgg_loss_2: 36.917198181152344
+INFO:__main__:vgg_loss_3: 26.672433853149414
+INFO:__main__:vgg_loss_4: 46.155879974365234
+INFO:__main__:vgg_loss_5: 1.549934983253479
+INFO:__main__:validation_loss: 2169.751708984375
+INFO:__main__:global_step: 46251
+INFO:__main__:kl_loss: 11624.9453125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1534.9998779296875
+INFO:__main__:loss: 1535.011474609375
+INFO:__main__:vgg_gram_loss_0: 7.248232364654541
+INFO:__main__:vgg_gram_loss_1: 21.65983772277832
+INFO:__main__:vgg_gram_loss_2: 22.797204971313477
+INFO:__main__:vgg_gram_loss_3: 18.71656608581543
+INFO:__main__:vgg_gram_loss_4: 74.11918640136719
+INFO:__main__:vgg_gram_loss_5: 0.15466660261154175
+INFO:__main__:vgg_loss_0: 11.791651725769043
+INFO:__main__:vgg_loss_1: 32.250301361083984
+INFO:__main__:vgg_loss_2: 40.416099548339844
+INFO:__main__:vgg_loss_3: 28.58140754699707
+INFO:__main__:vgg_loss_4: 47.765777587890625
+INFO:__main__:vgg_loss_5: 1.4990417957305908
+INFO:__main__:validation_loss: 1740.700439453125
+INFO:__main__:global_step: 46501
+INFO:__main__:kl_loss: 10672.705078125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1604.95458984375
+INFO:__main__:loss: 1604.9652099609375
+INFO:__main__:vgg_gram_loss_0: 5.847805500030518
+INFO:__main__:vgg_gram_loss_1: 21.89736557006836
+INFO:__main__:vgg_gram_loss_2: 21.4101619720459
+INFO:__main__:vgg_gram_loss_3: 18.585386276245117
+INFO:__main__:vgg_gram_loss_4: 76.29341125488281
+INFO:__main__:vgg_gram_loss_5: 0.18277041614055634
+INFO:__main__:vgg_loss_0: 12.595179557800293
+INFO:__main__:vgg_loss_1: 32.220523834228516
+INFO:__main__:vgg_loss_2: 45.066402435302734
+INFO:__main__:vgg_loss_3: 31.926246643066406
+INFO:__main__:vgg_loss_4: 53.212242126464844
+INFO:__main__:vgg_loss_5: 1.753411054611206
+INFO:__main__:validation_loss: 1601.732666015625
+INFO:__main__:global_step: 46751
+INFO:__main__:kl_loss: 9982.9345703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1473.312744140625
+INFO:__main__:loss: 1473.32275390625
+INFO:__main__:vgg_gram_loss_0: 6.299622535705566
+INFO:__main__:vgg_gram_loss_1: 19.18580436706543
+INFO:__main__:vgg_gram_loss_2: 18.69093132019043
+INFO:__main__:vgg_gram_loss_3: 15.918965339660645
+INFO:__main__:vgg_gram_loss_4: 72.23751831054688
+INFO:__main__:vgg_gram_loss_5: 0.16409669816493988
+INFO:__main__:vgg_loss_0: 12.512399673461914
+INFO:__main__:vgg_loss_1: 31.974864959716797
+INFO:__main__:vgg_loss_2: 40.401756286621094
+INFO:__main__:vgg_loss_3: 27.9616756439209
+INFO:__main__:vgg_loss_4: 47.72086715698242
+INFO:__main__:vgg_loss_5: 1.5940617322921753
+INFO:__main__:validation_loss: 2159.54052734375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 47001
+INFO:__main__:kl_loss: 10078.283203125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1431.392822265625
+INFO:__main__:loss: 1431.4029541015625
+INFO:__main__:vgg_gram_loss_0: 5.914727687835693
+INFO:__main__:vgg_gram_loss_1: 15.144320487976074
+INFO:__main__:vgg_gram_loss_2: 18.318933486938477
+INFO:__main__:vgg_gram_loss_3: 15.577391624450684
+INFO:__main__:vgg_gram_loss_4: 72.8405532836914
+INFO:__main__:vgg_gram_loss_5: 0.17846618592739105
+INFO:__main__:vgg_loss_0: 10.73085880279541
+INFO:__main__:vgg_loss_1: 26.31502914428711
+INFO:__main__:vgg_loss_2: 39.2459602355957
+INFO:__main__:vgg_loss_3: 28.666126251220703
+INFO:__main__:vgg_loss_4: 51.59365463256836
+INFO:__main__:vgg_loss_5: 1.7525335550308228
+INFO:__main__:validation_loss: 1978.1878662109375
+INFO:__main__:global_step: 47251
+INFO:__main__:kl_loss: 10986.896484375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1664.9908447265625
+INFO:__main__:loss: 1665.0018310546875
+INFO:__main__:vgg_gram_loss_0: 7.318727016448975
+INFO:__main__:vgg_gram_loss_1: 18.565624237060547
+INFO:__main__:vgg_gram_loss_2: 21.79892349243164
+INFO:__main__:vgg_gram_loss_3: 17.390148162841797
+INFO:__main__:vgg_gram_loss_4: 84.30352783203125
+INFO:__main__:vgg_gram_loss_5: 0.198177769780159
+INFO:__main__:vgg_loss_0: 12.391894340515137
+INFO:__main__:vgg_loss_1: 29.72643280029297
+INFO:__main__:vgg_loss_2: 45.29069137573242
+INFO:__main__:vgg_loss_3: 33.64618682861328
+INFO:__main__:vgg_loss_4: 60.37594985961914
+INFO:__main__:vgg_loss_5: 1.9918694496154785
+INFO:__main__:validation_loss: 1929.20556640625
+INFO:__main__:global_step: 47501
+INFO:__main__:kl_loss: 10182.921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1380.0665283203125
+INFO:__main__:loss: 1380.07666015625
+INFO:__main__:vgg_gram_loss_0: 5.493350028991699
+INFO:__main__:vgg_gram_loss_1: 21.50483512878418
+INFO:__main__:vgg_gram_loss_2: 21.4522762298584
+INFO:__main__:vgg_gram_loss_3: 15.535638809204102
+INFO:__main__:vgg_gram_loss_4: 65.71512603759766
+INFO:__main__:vgg_gram_loss_5: 0.14386455714702606
+INFO:__main__:vgg_loss_0: 10.470024108886719
+INFO:__main__:vgg_loss_1: 25.66710090637207
+INFO:__main__:vgg_loss_2: 36.51114273071289
+INFO:__main__:vgg_loss_3: 26.23501205444336
+INFO:__main__:vgg_loss_4: 45.74760818481445
+INFO:__main__:vgg_loss_5: 1.537335991859436
+INFO:__main__:validation_loss: 2030.8070068359375
+INFO:__main__:global_step: 47751
+INFO:__main__:kl_loss: 8652.5986328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1464.86181640625
+INFO:__main__:loss: 1464.8704833984375
+INFO:__main__:vgg_gram_loss_0: 11.339594841003418
+INFO:__main__:vgg_gram_loss_1: 18.35701560974121
+INFO:__main__:vgg_gram_loss_2: 20.269136428833008
+INFO:__main__:vgg_gram_loss_3: 17.011545181274414
+INFO:__main__:vgg_gram_loss_4: 75.84097290039062
+INFO:__main__:vgg_gram_loss_5: 0.16368991136550903
+INFO:__main__:vgg_loss_0: 10.44469165802002
+INFO:__main__:vgg_loss_1: 24.215784072875977
+INFO:__main__:vgg_loss_2: 35.9162483215332
+INFO:__main__:vgg_loss_3: 27.062015533447266
+INFO:__main__:vgg_loss_4: 50.63377380371094
+INFO:__main__:vgg_loss_5: 1.717882513999939
+INFO:__main__:validation_loss: 1622.1221923828125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 48001
+INFO:__main__:kl_loss: 10134.9765625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1321.98388671875
+INFO:__main__:loss: 1321.9940185546875
+INFO:__main__:vgg_gram_loss_0: 6.113168239593506
+INFO:__main__:vgg_gram_loss_1: 18.699430465698242
+INFO:__main__:vgg_gram_loss_2: 18.64750862121582
+INFO:__main__:vgg_gram_loss_3: 15.411559104919434
+INFO:__main__:vgg_gram_loss_4: 66.18915557861328
+INFO:__main__:vgg_gram_loss_5: 0.1562570333480835
+INFO:__main__:vgg_loss_0: 9.754739761352539
+INFO:__main__:vgg_loss_1: 24.668560028076172
+INFO:__main__:vgg_loss_2: 35.20296096801758
+INFO:__main__:vgg_loss_3: 24.725141525268555
+INFO:__main__:vgg_loss_4: 43.35619354248047
+INFO:__main__:vgg_loss_5: 1.4721070528030396
+INFO:__main__:validation_loss: 1860.442138671875
+INFO:__main__:global_step: 48251
+INFO:__main__:kl_loss: 10799.1328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1588.071533203125
+INFO:__main__:loss: 1588.082275390625
+INFO:__main__:vgg_gram_loss_0: 7.068974494934082
+INFO:__main__:vgg_gram_loss_1: 21.27472686767578
+INFO:__main__:vgg_gram_loss_2: 20.338542938232422
+INFO:__main__:vgg_gram_loss_3: 17.793970108032227
+INFO:__main__:vgg_gram_loss_4: 83.56456756591797
+INFO:__main__:vgg_gram_loss_5: 0.18094377219676971
+INFO:__main__:vgg_loss_0: 11.307072639465332
+INFO:__main__:vgg_loss_1: 27.885665893554688
+INFO:__main__:vgg_loss_2: 41.30362319946289
+INFO:__main__:vgg_loss_3: 30.19803237915039
+INFO:__main__:vgg_loss_4: 54.84920883178711
+INFO:__main__:vgg_loss_5: 1.8490036725997925
+INFO:__main__:validation_loss: 1754.48828125
+INFO:__main__:global_step: 48501
+INFO:__main__:kl_loss: 11141.1884765625
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1241.2001953125
+INFO:__main__:loss: 1241.2113037109375
+INFO:__main__:vgg_gram_loss_0: 6.524314880371094
+INFO:__main__:vgg_gram_loss_1: 14.145950317382812
+INFO:__main__:vgg_gram_loss_2: 19.864683151245117
+INFO:__main__:vgg_gram_loss_3: 14.449272155761719
+INFO:__main__:vgg_gram_loss_4: 62.964393615722656
+INFO:__main__:vgg_gram_loss_5: 0.1428176611661911
+INFO:__main__:vgg_loss_0: 8.853433609008789
+INFO:__main__:vgg_loss_1: 21.497922897338867
+INFO:__main__:vgg_loss_2: 32.32904815673828
+INFO:__main__:vgg_loss_3: 23.743886947631836
+INFO:__main__:vgg_loss_4: 42.31006622314453
+INFO:__main__:vgg_loss_5: 1.414239764213562
+INFO:__main__:validation_loss: 1661.0098876953125
+INFO:__main__:global_step: 48751
+INFO:__main__:kl_loss: 11161.685546875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1458.2374267578125
+INFO:__main__:loss: 1458.24853515625
+INFO:__main__:vgg_gram_loss_0: 8.194229125976562
+INFO:__main__:vgg_gram_loss_1: 21.387786865234375
+INFO:__main__:vgg_gram_loss_2: 19.230297088623047
+INFO:__main__:vgg_gram_loss_3: 17.19060707092285
+INFO:__main__:vgg_gram_loss_4: 73.0634536743164
+INFO:__main__:vgg_gram_loss_5: 0.15924127399921417
+INFO:__main__:vgg_loss_0: 11.185633659362793
+INFO:__main__:vgg_loss_1: 28.76934051513672
+INFO:__main__:vgg_loss_2: 36.9621467590332
+INFO:__main__:vgg_loss_3: 26.794952392578125
+INFO:__main__:vgg_loss_4: 47.13070297241211
+INFO:__main__:vgg_loss_5: 1.579094648361206
+INFO:__main__:validation_loss: 2132.970947265625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 49001
+INFO:__main__:kl_loss: 10595.41796875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1872.53076171875
+INFO:__main__:loss: 1872.5413818359375
+INFO:__main__:vgg_gram_loss_0: 6.234829902648926
+INFO:__main__:vgg_gram_loss_1: 28.839014053344727
+INFO:__main__:vgg_gram_loss_2: 26.18931007385254
+INFO:__main__:vgg_gram_loss_3: 21.980501174926758
+INFO:__main__:vgg_gram_loss_4: 86.63667297363281
+INFO:__main__:vgg_gram_loss_5: 0.19575807452201843
+INFO:__main__:vgg_loss_0: 15.651512145996094
+INFO:__main__:vgg_loss_1: 42.12035369873047
+INFO:__main__:vgg_loss_2: 50.76625061035156
+INFO:__main__:vgg_loss_3: 34.42153549194336
+INFO:__main__:vgg_loss_4: 59.507530212402344
+INFO:__main__:vgg_loss_5: 1.9628841876983643
+INFO:__main__:validation_loss: 1724.3360595703125
+INFO:__main__:global_step: 49251
+INFO:__main__:kl_loss: 9937.1220703125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1278.438232421875
+INFO:__main__:loss: 1278.4481201171875
+INFO:__main__:vgg_gram_loss_0: 7.141271114349365
+INFO:__main__:vgg_gram_loss_1: 14.809605598449707
+INFO:__main__:vgg_gram_loss_2: 16.92582130432129
+INFO:__main__:vgg_gram_loss_3: 13.93735408782959
+INFO:__main__:vgg_gram_loss_4: 64.91397094726562
+INFO:__main__:vgg_gram_loss_5: 0.15024583041667938
+INFO:__main__:vgg_loss_0: 9.361444473266602
+INFO:__main__:vgg_loss_1: 23.374597549438477
+INFO:__main__:vgg_loss_2: 34.301395416259766
+INFO:__main__:vgg_loss_3: 25.015689849853516
+INFO:__main__:vgg_loss_4: 44.26148223876953
+INFO:__main__:vgg_loss_5: 1.4947830438613892
+INFO:__main__:validation_loss: 1633.2581787109375
+INFO:__main__:global_step: 49501
+INFO:__main__:kl_loss: 11633.6611328125
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1451.303466796875
+INFO:__main__:loss: 1451.3150634765625
+INFO:__main__:vgg_gram_loss_0: 5.764016628265381
+INFO:__main__:vgg_gram_loss_1: 14.599926948547363
+INFO:__main__:vgg_gram_loss_2: 22.467702865600586
+INFO:__main__:vgg_gram_loss_3: 17.43221664428711
+INFO:__main__:vgg_gram_loss_4: 78.71646881103516
+INFO:__main__:vgg_gram_loss_5: 0.1577153354883194
+INFO:__main__:vgg_loss_0: 11.183650016784668
+INFO:__main__:vgg_loss_1: 27.313695907592773
+INFO:__main__:vgg_loss_2: 37.94483184814453
+INFO:__main__:vgg_loss_3: 26.80497169494629
+INFO:__main__:vgg_loss_4: 46.3929328918457
+INFO:__main__:vgg_loss_5: 1.482581615447998
+INFO:__main__:validation_loss: 1969.69921875
+INFO:__main__:global_step: 49751
+INFO:__main__:kl_loss: 10819.669921875
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1361.5533447265625
+INFO:__main__:loss: 1361.564208984375
+INFO:__main__:vgg_gram_loss_0: 6.355626583099365
+INFO:__main__:vgg_gram_loss_1: 11.874571800231934
+INFO:__main__:vgg_gram_loss_2: 17.2707462310791
+INFO:__main__:vgg_gram_loss_3: 12.773699760437012
+INFO:__main__:vgg_gram_loss_4: 66.23451232910156
+INFO:__main__:vgg_gram_loss_5: 0.16165654361248016
+INFO:__main__:vgg_loss_0: 11.160950660705566
+INFO:__main__:vgg_loss_1: 24.960468292236328
+INFO:__main__:vgg_loss_2: 38.56708526611328
+INFO:__main__:vgg_loss_3: 28.930692672729492
+INFO:__main__:vgg_loss_4: 52.32777404785156
+INFO:__main__:vgg_loss_5: 1.6928824186325073
+INFO:__main__:validation_loss: 1607.077880859375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 50001
+INFO:__main__:kl_loss: 10482.8349609375
+INFO:__main__:kl_weight: 9.999999974752427e-07
+INFO:__main__:likelihood_loss: 1497.994384765625
+INFO:__main__:loss: 1498.0048828125
+INFO:__main__:vgg_gram_loss_0: 6.353371620178223
+INFO:__main__:vgg_gram_loss_1: 16.795412063598633
+INFO:__main__:vgg_gram_loss_2: 21.50495719909668
+INFO:__main__:vgg_gram_loss_3: 16.367414474487305
+INFO:__main__:vgg_gram_loss_4: 77.75798034667969
+INFO:__main__:vgg_gram_loss_5: 0.18754617869853973
+INFO:__main__:vgg_loss_0: 10.910284042358398
+INFO:__main__:vgg_loss_1: 27.008249282836914
+INFO:__main__:vgg_loss_2: 40.579803466796875
+INFO:__main__:vgg_loss_3: 29.356081008911133
+INFO:__main__:vgg_loss_4: 51.09076690673828
+INFO:__main__:vgg_loss_5: 1.6869752407073975
+INFO:__main__:validation_loss: 1919.357177734375
+INFO:__main__:global_step: 50251
+INFO:__main__:kl_loss: 6441.45703125
+INFO:__main__:kl_weight: 0.010000989772379398
+INFO:__main__:likelihood_loss: 1588.80517578125
+INFO:__main__:loss: 1653.22607421875
+INFO:__main__:vgg_gram_loss_0: 11.603306770324707
+INFO:__main__:vgg_gram_loss_1: 22.275760650634766
+INFO:__main__:vgg_gram_loss_2: 23.075563430786133
+INFO:__main__:vgg_gram_loss_3: 18.30158042907715
+INFO:__main__:vgg_gram_loss_4: 77.6226577758789
+INFO:__main__:vgg_gram_loss_5: 0.16284334659576416
+INFO:__main__:vgg_loss_0: 12.071303367614746
+INFO:__main__:vgg_loss_1: 29.309852600097656
+INFO:__main__:vgg_loss_2: 41.47458267211914
+INFO:__main__:vgg_loss_3: 29.454029083251953
+INFO:__main__:vgg_loss_4: 50.73735046386719
+INFO:__main__:vgg_loss_5: 1.6722099781036377
+INFO:__main__:validation_loss: 2064.455078125
+INFO:__main__:global_step: 50501
+INFO:__main__:kl_loss: 3479.3857421875
+INFO:__main__:kl_weight: 0.02000097930431366
+INFO:__main__:likelihood_loss: 1429.242431640625
+INFO:__main__:loss: 1498.83349609375
+INFO:__main__:vgg_gram_loss_0: 7.872462749481201
+INFO:__main__:vgg_gram_loss_1: 15.969047546386719
+INFO:__main__:vgg_gram_loss_2: 17.567533493041992
+INFO:__main__:vgg_gram_loss_3: 13.820124626159668
+INFO:__main__:vgg_gram_loss_4: 70.57093048095703
+INFO:__main__:vgg_gram_loss_5: 0.18946531414985657
+INFO:__main__:vgg_loss_0: 10.881997108459473
+INFO:__main__:vgg_loss_1: 27.785066604614258
+INFO:__main__:vgg_loss_2: 40.48052215576172
+INFO:__main__:vgg_loss_3: 28.765846252441406
+INFO:__main__:vgg_loss_4: 50.20218276977539
+INFO:__main__:vgg_loss_5: 1.7433021068572998
+INFO:__main__:validation_loss: 1751.7431640625
+INFO:__main__:global_step: 50751
+INFO:__main__:kl_loss: 2181.4326171875
+INFO:__main__:kl_weight: 0.030000969767570496
+INFO:__main__:likelihood_loss: 1419.87939453125
+INFO:__main__:loss: 1485.324462890625
+INFO:__main__:vgg_gram_loss_0: 7.737094402313232
+INFO:__main__:vgg_gram_loss_1: 18.276548385620117
+INFO:__main__:vgg_gram_loss_2: 18.63458824157715
+INFO:__main__:vgg_gram_loss_3: 14.927560806274414
+INFO:__main__:vgg_gram_loss_4: 66.627685546875
+INFO:__main__:vgg_gram_loss_5: 0.15339207649230957
+INFO:__main__:vgg_loss_0: 11.689186096191406
+INFO:__main__:vgg_loss_1: 28.27732276916504
+INFO:__main__:vgg_loss_2: 39.37963104248047
+INFO:__main__:vgg_loss_3: 28.145734786987305
+INFO:__main__:vgg_loss_4: 48.5289306640625
+INFO:__main__:vgg_loss_5: 1.5982006788253784
+INFO:__main__:validation_loss: 1976.7685546875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 51001
+INFO:__main__:kl_loss: 2096.58837890625
+INFO:__main__:kl_weight: 0.04000095650553703
+INFO:__main__:likelihood_loss: 1863.44091796875
+INFO:__main__:loss: 1947.306396484375
+INFO:__main__:vgg_gram_loss_0: 5.469057559967041
+INFO:__main__:vgg_gram_loss_1: 23.44309425354004
+INFO:__main__:vgg_gram_loss_2: 32.70433044433594
+INFO:__main__:vgg_gram_loss_3: 24.84476661682129
+INFO:__main__:vgg_gram_loss_4: 92.6097640991211
+INFO:__main__:vgg_gram_loss_5: 0.18744021654129028
+INFO:__main__:vgg_loss_0: 13.651097297668457
+INFO:__main__:vgg_loss_1: 36.333900451660156
+INFO:__main__:vgg_loss_2: 49.459991455078125
+INFO:__main__:vgg_loss_3: 34.59016799926758
+INFO:__main__:vgg_loss_4: 57.5623664855957
+INFO:__main__:vgg_loss_5: 1.8322093486785889
+INFO:__main__:validation_loss: 2109.63525390625
+INFO:__main__:global_step: 51251
+INFO:__main__:kl_loss: 1139.9891357421875
+INFO:__main__:kl_weight: 0.05000094696879387
+INFO:__main__:likelihood_loss: 1247.5076904296875
+INFO:__main__:loss: 1304.5081787109375
+INFO:__main__:vgg_gram_loss_0: 7.700303077697754
+INFO:__main__:vgg_gram_loss_1: 13.78819465637207
+INFO:__main__:vgg_gram_loss_2: 17.027082443237305
+INFO:__main__:vgg_gram_loss_3: 14.051159858703613
+INFO:__main__:vgg_gram_loss_4: 65.38643646240234
+INFO:__main__:vgg_gram_loss_5: 0.13996851444244385
+INFO:__main__:vgg_loss_0: 8.920395851135254
+INFO:__main__:vgg_loss_1: 21.259119033813477
+INFO:__main__:vgg_loss_2: 32.462154388427734
+INFO:__main__:vgg_loss_3: 24.015968322753906
+INFO:__main__:vgg_loss_4: 43.305912017822266
+INFO:__main__:vgg_loss_5: 1.4448519945144653
+INFO:__main__:validation_loss: 2189.865966796875
+INFO:__main__:global_step: 51501
+INFO:__main__:kl_loss: 1096.5625
+INFO:__main__:kl_weight: 0.060000937432050705
+INFO:__main__:likelihood_loss: 1434.164306640625
+INFO:__main__:loss: 1499.9591064453125
+INFO:__main__:vgg_gram_loss_0: 5.695139408111572
+INFO:__main__:vgg_gram_loss_1: 19.66288948059082
+INFO:__main__:vgg_gram_loss_2: 19.621091842651367
+INFO:__main__:vgg_gram_loss_3: 16.523672103881836
+INFO:__main__:vgg_gram_loss_4: 70.08119201660156
+INFO:__main__:vgg_gram_loss_5: 0.15740208327770233
+INFO:__main__:vgg_loss_0: 11.443211555480957
+INFO:__main__:vgg_loss_1: 28.306964874267578
+INFO:__main__:vgg_loss_2: 37.93976593017578
+INFO:__main__:vgg_loss_3: 27.58085823059082
+INFO:__main__:vgg_loss_4: 48.209373474121094
+INFO:__main__:vgg_loss_5: 1.611291766166687
+INFO:__main__:validation_loss: 1622.193359375
+INFO:__main__:global_step: 51751
+INFO:__main__:kl_loss: 974.5350952148438
+INFO:__main__:kl_weight: 0.07000092417001724
+INFO:__main__:likelihood_loss: 1551.93701171875
+INFO:__main__:loss: 1620.1553955078125
+INFO:__main__:vgg_gram_loss_0: 8.096752166748047
+INFO:__main__:vgg_gram_loss_1: 17.85909080505371
+INFO:__main__:vgg_gram_loss_2: 21.68777847290039
+INFO:__main__:vgg_gram_loss_3: 16.98805809020996
+INFO:__main__:vgg_gram_loss_4: 78.21915435791016
+INFO:__main__:vgg_gram_loss_5: 0.18453454971313477
+INFO:__main__:vgg_loss_0: 11.632030487060547
+INFO:__main__:vgg_loss_1: 30.164386749267578
+INFO:__main__:vgg_loss_2: 42.687538146972656
+INFO:__main__:vgg_loss_3: 29.92292594909668
+INFO:__main__:vgg_loss_4: 51.24711227416992
+INFO:__main__:vgg_loss_5: 1.6980128288269043
+INFO:__main__:validation_loss: 1853.6083984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 52001
+INFO:__main__:kl_loss: 796.4420166015625
+INFO:__main__:kl_weight: 0.08000091463327408
+INFO:__main__:likelihood_loss: 1652.8472900390625
+INFO:__main__:loss: 1716.5633544921875
+INFO:__main__:vgg_gram_loss_0: 5.346341133117676
+INFO:__main__:vgg_gram_loss_1: 18.816364288330078
+INFO:__main__:vgg_gram_loss_2: 26.21480369567871
+INFO:__main__:vgg_gram_loss_3: 21.443639755249023
+INFO:__main__:vgg_gram_loss_4: 85.23721313476562
+INFO:__main__:vgg_gram_loss_5: 0.1823924034833908
+INFO:__main__:vgg_loss_0: 11.99412727355957
+INFO:__main__:vgg_loss_1: 31.607637405395508
+INFO:__main__:vgg_loss_2: 44.40090560913086
+INFO:__main__:vgg_loss_3: 31.070636749267578
+INFO:__main__:vgg_loss_4: 52.53947448730469
+INFO:__main__:vgg_loss_5: 1.7159429788589478
+INFO:__main__:validation_loss: 1907.5570068359375
+INFO:__main__:global_step: 52251
+INFO:__main__:kl_loss: 743.0657348632812
+INFO:__main__:kl_weight: 0.09000090509653091
+INFO:__main__:likelihood_loss: 1385.849365234375
+INFO:__main__:loss: 1452.7259521484375
+INFO:__main__:vgg_gram_loss_0: 5.245667934417725
+INFO:__main__:vgg_gram_loss_1: 13.94110107421875
+INFO:__main__:vgg_gram_loss_2: 20.320175170898438
+INFO:__main__:vgg_gram_loss_3: 15.484875679016113
+INFO:__main__:vgg_gram_loss_4: 71.19024658203125
+INFO:__main__:vgg_gram_loss_5: 0.15179328620433807
+INFO:__main__:vgg_loss_0: 10.517475128173828
+INFO:__main__:vgg_loss_1: 24.974077224731445
+INFO:__main__:vgg_loss_2: 36.866058349609375
+INFO:__main__:vgg_loss_3: 27.30190658569336
+INFO:__main__:vgg_loss_4: 49.605194091796875
+INFO:__main__:vgg_loss_5: 1.5712664127349854
+INFO:__main__:validation_loss: 2022.8323974609375
+INFO:__main__:global_step: 52501
+INFO:__main__:kl_loss: 637.813232421875
+INFO:__main__:kl_weight: 0.10000089555978775
+INFO:__main__:likelihood_loss: 1760.822021484375
+INFO:__main__:loss: 1824.6038818359375
+INFO:__main__:vgg_gram_loss_0: 5.313925266265869
+INFO:__main__:vgg_gram_loss_1: 21.3103084564209
+INFO:__main__:vgg_gram_loss_2: 29.3493595123291
+INFO:__main__:vgg_gram_loss_3: 20.926727294921875
+INFO:__main__:vgg_gram_loss_4: 85.4075927734375
+INFO:__main__:vgg_gram_loss_5: 0.18467698991298676
+INFO:__main__:vgg_loss_0: 13.07962417602539
+INFO:__main__:vgg_loss_1: 32.572757720947266
+INFO:__main__:vgg_loss_2: 49.33335494995117
+INFO:__main__:vgg_loss_3: 34.4154167175293
+INFO:__main__:vgg_loss_4: 58.431766510009766
+INFO:__main__:vgg_loss_5: 1.8388787508010864
+INFO:__main__:validation_loss: 1809.31005859375
+INFO:__main__:global_step: 52751
+INFO:__main__:kl_loss: 620.4498291015625
+INFO:__main__:kl_weight: 0.11000088602304459
+INFO:__main__:likelihood_loss: 1946.52197265625
+INFO:__main__:loss: 2014.77197265625
+INFO:__main__:vgg_gram_loss_0: 6.0110602378845215
+INFO:__main__:vgg_gram_loss_1: 39.03408432006836
+INFO:__main__:vgg_gram_loss_2: 27.813644409179688
+INFO:__main__:vgg_gram_loss_3: 24.762374877929688
+INFO:__main__:vgg_gram_loss_4: 95.22132110595703
+INFO:__main__:vgg_gram_loss_5: 0.2029167264699936
+INFO:__main__:vgg_loss_0: 14.952186584472656
+INFO:__main__:vgg_loss_1: 39.09959411621094
+INFO:__main__:vgg_loss_2: 47.42322540283203
+INFO:__main__:vgg_loss_3: 34.21035385131836
+INFO:__main__:vgg_loss_4: 58.67253112792969
+INFO:__main__:vgg_loss_5: 1.9010875225067139
+INFO:__main__:validation_loss: 1947.168212890625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 53001
+INFO:__main__:kl_loss: 500.6832275390625
+INFO:__main__:kl_weight: 0.12000087648630142
+INFO:__main__:likelihood_loss: 1537.3515625
+INFO:__main__:loss: 1597.4339599609375
+INFO:__main__:vgg_gram_loss_0: 11.605873107910156
+INFO:__main__:vgg_gram_loss_1: 17.331052780151367
+INFO:__main__:vgg_gram_loss_2: 23.5655574798584
+INFO:__main__:vgg_gram_loss_3: 17.11884117126465
+INFO:__main__:vgg_gram_loss_4: 74.2616958618164
+INFO:__main__:vgg_gram_loss_5: 0.16537658870220184
+INFO:__main__:vgg_loss_0: 11.59255313873291
+INFO:__main__:vgg_loss_1: 27.062543869018555
+INFO:__main__:vgg_loss_2: 41.53169250488281
+INFO:__main__:vgg_loss_3: 29.53998565673828
+INFO:__main__:vgg_loss_4: 52.00334548950195
+INFO:__main__:vgg_loss_5: 1.6917881965637207
+INFO:__main__:validation_loss: 2117.257080078125
+INFO:__main__:global_step: 53251
+INFO:__main__:kl_loss: 482.48944091796875
+INFO:__main__:kl_weight: 0.13000085949897766
+INFO:__main__:likelihood_loss: 1404.3759765625
+INFO:__main__:loss: 1467.0999755859375
+INFO:__main__:vgg_gram_loss_0: 7.9877848625183105
+INFO:__main__:vgg_gram_loss_1: 14.766952514648438
+INFO:__main__:vgg_gram_loss_2: 18.378337860107422
+INFO:__main__:vgg_gram_loss_3: 16.32552719116211
+INFO:__main__:vgg_gram_loss_4: 70.78013610839844
+INFO:__main__:vgg_gram_loss_5: 0.1577749103307724
+INFO:__main__:vgg_loss_0: 11.03810977935791
+INFO:__main__:vgg_loss_1: 27.573923110961914
+INFO:__main__:vgg_loss_2: 38.46590042114258
+INFO:__main__:vgg_loss_3: 26.80157470703125
+INFO:__main__:vgg_loss_4: 47.02141189575195
+INFO:__main__:vgg_loss_5: 1.5777209997177124
+INFO:__main__:validation_loss: 1684.718994140625
+INFO:__main__:global_step: 53501
+INFO:__main__:kl_loss: 461.0302429199219
+INFO:__main__:kl_weight: 0.1400008499622345
+INFO:__main__:likelihood_loss: 1447.0191650390625
+INFO:__main__:loss: 1511.5638427734375
+INFO:__main__:vgg_gram_loss_0: 4.7887396812438965
+INFO:__main__:vgg_gram_loss_1: 14.607309341430664
+INFO:__main__:vgg_gram_loss_2: 19.768455505371094
+INFO:__main__:vgg_gram_loss_3: 15.794618606567383
+INFO:__main__:vgg_gram_loss_4: 71.68677520751953
+INFO:__main__:vgg_gram_loss_5: 0.16718849539756775
+INFO:__main__:vgg_loss_0: 11.967134475708008
+INFO:__main__:vgg_loss_1: 28.10008430480957
+INFO:__main__:vgg_loss_2: 40.48341751098633
+INFO:__main__:vgg_loss_3: 29.450363159179688
+INFO:__main__:vgg_loss_4: 50.92267990112305
+INFO:__main__:vgg_loss_5: 1.6670780181884766
+INFO:__main__:validation_loss: 2204.85302734375
+INFO:__main__:global_step: 53751
+INFO:__main__:kl_loss: 431.852294921875
+INFO:__main__:kl_weight: 0.15000084042549133
+INFO:__main__:likelihood_loss: 1411.210693359375
+INFO:__main__:loss: 1475.9888916015625
+INFO:__main__:vgg_gram_loss_0: 4.401966571807861
+INFO:__main__:vgg_gram_loss_1: 16.503515243530273
+INFO:__main__:vgg_gram_loss_2: 18.363956451416016
+INFO:__main__:vgg_gram_loss_3: 15.213739395141602
+INFO:__main__:vgg_gram_loss_4: 69.20381164550781
+INFO:__main__:vgg_gram_loss_5: 0.15624019503593445
+INFO:__main__:vgg_loss_0: 11.972630500793457
+INFO:__main__:vgg_loss_1: 29.638750076293945
+INFO:__main__:vgg_loss_2: 39.531620025634766
+INFO:__main__:vgg_loss_3: 28.06954574584961
+INFO:__main__:vgg_loss_4: 47.68438720703125
+INFO:__main__:vgg_loss_5: 1.5019475221633911
+INFO:__main__:validation_loss: 1901.4234619140625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 54001
+INFO:__main__:kl_loss: 462.9907531738281
+INFO:__main__:kl_weight: 0.16000083088874817
+INFO:__main__:likelihood_loss: 1842.2979736328125
+INFO:__main__:loss: 1916.3768310546875
+INFO:__main__:vgg_gram_loss_0: 3.738231658935547
+INFO:__main__:vgg_gram_loss_1: 23.489931106567383
+INFO:__main__:vgg_gram_loss_2: 29.33441162109375
+INFO:__main__:vgg_gram_loss_3: 21.77472686767578
+INFO:__main__:vgg_gram_loss_4: 91.37846374511719
+INFO:__main__:vgg_gram_loss_5: 0.2116180658340454
+INFO:__main__:vgg_loss_0: 14.744144439697266
+INFO:__main__:vgg_loss_1: 38.0261344909668
+INFO:__main__:vgg_loss_2: 52.06601333618164
+INFO:__main__:vgg_loss_3: 34.94027328491211
+INFO:__main__:vgg_loss_4: 56.85561752319336
+INFO:__main__:vgg_loss_5: 1.900015115737915
+INFO:__main__:validation_loss: 2100.53955078125
+INFO:__main__:global_step: 54251
+INFO:__main__:kl_loss: 358.58038330078125
+INFO:__main__:kl_weight: 0.170000821352005
+INFO:__main__:likelihood_loss: 1339.577880859375
+INFO:__main__:loss: 1400.536865234375
+INFO:__main__:vgg_gram_loss_0: 3.810626745223999
+INFO:__main__:vgg_gram_loss_1: 13.236065864562988
+INFO:__main__:vgg_gram_loss_2: 17.399415969848633
+INFO:__main__:vgg_gram_loss_3: 14.694482803344727
+INFO:__main__:vgg_gram_loss_4: 69.71311950683594
+INFO:__main__:vgg_gram_loss_5: 0.17458117008209229
+INFO:__main__:vgg_loss_0: 10.67831802368164
+INFO:__main__:vgg_loss_1: 26.348798751831055
+INFO:__main__:vgg_loss_2: 38.10749816894531
+INFO:__main__:vgg_loss_3: 26.664831161499023
+INFO:__main__:vgg_loss_4: 45.56396484375
+INFO:__main__:vgg_loss_5: 1.5238978862762451
+INFO:__main__:validation_loss: 1801.173828125
+INFO:__main__:global_step: 54501
+INFO:__main__:kl_loss: 309.11846923828125
+INFO:__main__:kl_weight: 0.18000081181526184
+INFO:__main__:likelihood_loss: 1565.0927734375
+INFO:__main__:loss: 1620.734375
+INFO:__main__:vgg_gram_loss_0: 6.526357173919678
+INFO:__main__:vgg_gram_loss_1: 19.692264556884766
+INFO:__main__:vgg_gram_loss_2: 20.819067001342773
+INFO:__main__:vgg_gram_loss_3: 18.053436279296875
+INFO:__main__:vgg_gram_loss_4: 79.7742919921875
+INFO:__main__:vgg_gram_loss_5: 0.1723712980747223
+INFO:__main__:vgg_loss_0: 13.056347846984863
+INFO:__main__:vgg_loss_1: 30.305248260498047
+INFO:__main__:vgg_loss_2: 41.51729965209961
+INFO:__main__:vgg_loss_3: 29.634963989257812
+INFO:__main__:vgg_loss_4: 51.80848693847656
+INFO:__main__:vgg_loss_5: 1.6584300994873047
+INFO:__main__:validation_loss: 1592.962158203125
+INFO:__main__:global_step: 54751
+INFO:__main__:kl_loss: 355.88592529296875
+INFO:__main__:kl_weight: 0.19000080227851868
+INFO:__main__:likelihood_loss: 1548.958984375
+INFO:__main__:loss: 1616.57763671875
+INFO:__main__:vgg_gram_loss_0: 6.070631504058838
+INFO:__main__:vgg_gram_loss_1: 17.057086944580078
+INFO:__main__:vgg_gram_loss_2: 22.516061782836914
+INFO:__main__:vgg_gram_loss_3: 18.91379165649414
+INFO:__main__:vgg_gram_loss_4: 76.80543518066406
+INFO:__main__:vgg_gram_loss_5: 0.16906532645225525
+INFO:__main__:vgg_loss_0: 13.024688720703125
+INFO:__main__:vgg_loss_1: 30.129215240478516
+INFO:__main__:vgg_loss_2: 41.8564567565918
+INFO:__main__:vgg_loss_3: 29.795534133911133
+INFO:__main__:vgg_loss_4: 51.76190185546875
+INFO:__main__:vgg_loss_5: 1.691930890083313
+INFO:__main__:validation_loss: 1913.247802734375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 55001
+INFO:__main__:kl_loss: 322.4005432128906
+INFO:__main__:kl_weight: 0.2000007927417755
+INFO:__main__:likelihood_loss: 1534.005859375
+INFO:__main__:loss: 1598.4862060546875
+INFO:__main__:vgg_gram_loss_0: 10.427729606628418
+INFO:__main__:vgg_gram_loss_1: 17.52162742614746
+INFO:__main__:vgg_gram_loss_2: 22.398324966430664
+INFO:__main__:vgg_gram_loss_3: 18.782310485839844
+INFO:__main__:vgg_gram_loss_4: 79.3458251953125
+INFO:__main__:vgg_gram_loss_5: 0.16543404757976532
+INFO:__main__:vgg_loss_0: 12.260175704956055
+INFO:__main__:vgg_loss_1: 29.027027130126953
+INFO:__main__:vgg_loss_2: 38.96953201293945
+INFO:__main__:vgg_loss_3: 27.939149856567383
+INFO:__main__:vgg_loss_4: 48.373592376708984
+INFO:__main__:vgg_loss_5: 1.5904537439346313
+INFO:__main__:validation_loss: 2030.5093994140625
+INFO:__main__:global_step: 55251
+INFO:__main__:kl_loss: 366.66400146484375
+INFO:__main__:kl_weight: 0.21000078320503235
+INFO:__main__:likelihood_loss: 1675.959716796875
+INFO:__main__:loss: 1752.95947265625
+INFO:__main__:vgg_gram_loss_0: 7.035398960113525
+INFO:__main__:vgg_gram_loss_1: 20.947263717651367
+INFO:__main__:vgg_gram_loss_2: 24.243879318237305
+INFO:__main__:vgg_gram_loss_3: 20.27840232849121
+INFO:__main__:vgg_gram_loss_4: 88.2254409790039
+INFO:__main__:vgg_gram_loss_5: 0.2016974687576294
+INFO:__main__:vgg_loss_0: 11.855454444885254
+INFO:__main__:vgg_loss_1: 29.761552810668945
+INFO:__main__:vgg_loss_2: 43.730464935302734
+INFO:__main__:vgg_loss_3: 31.802310943603516
+INFO:__main__:vgg_loss_4: 55.30104446411133
+INFO:__main__:vgg_loss_5: 1.809045672416687
+INFO:__main__:validation_loss: 2055.14794921875
+INFO:__main__:global_step: 55501
+INFO:__main__:kl_loss: 295.6310729980469
+INFO:__main__:kl_weight: 0.22000077366828918
+INFO:__main__:likelihood_loss: 1755.7501220703125
+INFO:__main__:loss: 1820.7891845703125
+INFO:__main__:vgg_gram_loss_0: 10.51201343536377
+INFO:__main__:vgg_gram_loss_1: 27.488889694213867
+INFO:__main__:vgg_gram_loss_2: 26.922956466674805
+INFO:__main__:vgg_gram_loss_3: 21.056529998779297
+INFO:__main__:vgg_gram_loss_4: 85.41593170166016
+INFO:__main__:vgg_gram_loss_5: 0.19088390469551086
+INFO:__main__:vgg_loss_0: 14.30600643157959
+INFO:__main__:vgg_loss_1: 32.54475784301758
+INFO:__main__:vgg_loss_2: 44.064945220947266
+INFO:__main__:vgg_loss_3: 31.656850814819336
+INFO:__main__:vgg_loss_4: 55.168460845947266
+INFO:__main__:vgg_loss_5: 1.8217823505401611
+INFO:__main__:validation_loss: 1617.594970703125
+INFO:__main__:global_step: 55751
+INFO:__main__:kl_loss: 289.1202392578125
+INFO:__main__:kl_weight: 0.23000076413154602
+INFO:__main__:likelihood_loss: 1424.61474609375
+INFO:__main__:loss: 1491.1126708984375
+INFO:__main__:vgg_gram_loss_0: 4.744224548339844
+INFO:__main__:vgg_gram_loss_1: 16.211872100830078
+INFO:__main__:vgg_gram_loss_2: 17.896440505981445
+INFO:__main__:vgg_gram_loss_3: 14.85644245147705
+INFO:__main__:vgg_gram_loss_4: 73.3225326538086
+INFO:__main__:vgg_gram_loss_5: 0.16847817599773407
+INFO:__main__:vgg_loss_0: 11.480287551879883
+INFO:__main__:vgg_loss_1: 27.546110153198242
+INFO:__main__:vgg_loss_2: 38.94001770019531
+INFO:__main__:vgg_loss_3: 28.381101608276367
+INFO:__main__:vgg_loss_4: 49.70298385620117
+INFO:__main__:vgg_loss_5: 1.67245614528656
+INFO:__main__:validation_loss: 1813.496826171875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 56001
+INFO:__main__:kl_loss: 308.44256591796875
+INFO:__main__:kl_weight: 0.24000075459480286
+INFO:__main__:likelihood_loss: 1839.719482421875
+INFO:__main__:loss: 1913.7459716796875
+INFO:__main__:vgg_gram_loss_0: 6.148536205291748
+INFO:__main__:vgg_gram_loss_1: 44.439029693603516
+INFO:__main__:vgg_gram_loss_2: 30.517669677734375
+INFO:__main__:vgg_gram_loss_3: 25.989551544189453
+INFO:__main__:vgg_gram_loss_4: 83.1806411743164
+INFO:__main__:vgg_gram_loss_5: 0.1638626903295517
+INFO:__main__:vgg_loss_0: 14.204683303833008
+INFO:__main__:vgg_loss_1: 39.07451629638672
+INFO:__main__:vgg_loss_2: 42.76865005493164
+INFO:__main__:vgg_loss_3: 29.527097702026367
+INFO:__main__:vgg_loss_4: 50.26337814331055
+INFO:__main__:vgg_loss_5: 1.6662834882736206
+INFO:__main__:validation_loss: 1974.0245361328125
+INFO:__main__:global_step: 56251
+INFO:__main__:kl_loss: 250.94212341308594
+INFO:__main__:kl_weight: 0.2500007450580597
+INFO:__main__:likelihood_loss: 1559.380859375
+INFO:__main__:loss: 1622.1165771484375
+INFO:__main__:vgg_gram_loss_0: 7.549241542816162
+INFO:__main__:vgg_gram_loss_1: 18.741119384765625
+INFO:__main__:vgg_gram_loss_2: 20.824504852294922
+INFO:__main__:vgg_gram_loss_3: 16.466163635253906
+INFO:__main__:vgg_gram_loss_4: 77.28612518310547
+INFO:__main__:vgg_gram_loss_5: 0.18312078714370728
+INFO:__main__:vgg_loss_0: 11.924654960632324
+INFO:__main__:vgg_loss_1: 29.507802963256836
+INFO:__main__:vgg_loss_2: 42.73615264892578
+INFO:__main__:vgg_loss_3: 30.59701919555664
+INFO:__main__:vgg_loss_4: 54.2194938659668
+INFO:__main__:vgg_loss_5: 1.840773105621338
+INFO:__main__:validation_loss: 2466.46435546875
+INFO:__main__:global_step: 56501
+INFO:__main__:kl_loss: 244.99449157714844
+INFO:__main__:kl_weight: 0.26000073552131653
+INFO:__main__:likelihood_loss: 1746.5936279296875
+INFO:__main__:loss: 1810.2923583984375
+INFO:__main__:vgg_gram_loss_0: 13.306136131286621
+INFO:__main__:vgg_gram_loss_1: 34.73527145385742
+INFO:__main__:vgg_gram_loss_2: 28.598623275756836
+INFO:__main__:vgg_gram_loss_3: 22.409894943237305
+INFO:__main__:vgg_gram_loss_4: 87.85713958740234
+INFO:__main__:vgg_gram_loss_5: 0.15691962838172913
+INFO:__main__:vgg_loss_0: 11.87444019317627
+INFO:__main__:vgg_loss_1: 28.48001480102539
+INFO:__main__:vgg_loss_2: 39.46847915649414
+INFO:__main__:vgg_loss_3: 28.913679122924805
+INFO:__main__:vgg_loss_4: 51.895050048828125
+INFO:__main__:vgg_loss_5: 1.6230523586273193
+INFO:__main__:validation_loss: 2071.701416015625
+INFO:__main__:global_step: 56751
+INFO:__main__:kl_loss: 257.2318115234375
+INFO:__main__:kl_weight: 0.27000072598457336
+INFO:__main__:likelihood_loss: 1523.1878662109375
+INFO:__main__:loss: 1592.640625
+INFO:__main__:vgg_gram_loss_0: 7.933885097503662
+INFO:__main__:vgg_gram_loss_1: 17.093976974487305
+INFO:__main__:vgg_gram_loss_2: 19.91901206970215
+INFO:__main__:vgg_gram_loss_3: 16.41315460205078
+INFO:__main__:vgg_gram_loss_4: 76.2152328491211
+INFO:__main__:vgg_gram_loss_5: 0.19090986251831055
+INFO:__main__:vgg_loss_0: 10.965791702270508
+INFO:__main__:vgg_loss_1: 28.365589141845703
+INFO:__main__:vgg_loss_2: 41.338279724121094
+INFO:__main__:vgg_loss_3: 29.8920955657959
+INFO:__main__:vgg_loss_4: 54.499595642089844
+INFO:__main__:vgg_loss_5: 1.8100500106811523
+INFO:__main__:validation_loss: 1678.0438232421875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 57001
+INFO:__main__:kl_loss: 254.40325927734375
+INFO:__main__:kl_weight: 0.2800007164478302
+INFO:__main__:likelihood_loss: 1580.7952880859375
+INFO:__main__:loss: 1652.0283203125
+INFO:__main__:vgg_gram_loss_0: 5.988608360290527
+INFO:__main__:vgg_gram_loss_1: 22.908893585205078
+INFO:__main__:vgg_gram_loss_2: 24.294221878051758
+INFO:__main__:vgg_gram_loss_3: 18.163793563842773
+INFO:__main__:vgg_gram_loss_4: 76.63917541503906
+INFO:__main__:vgg_gram_loss_5: 0.18036030232906342
+INFO:__main__:vgg_loss_0: 12.071866989135742
+INFO:__main__:vgg_loss_1: 28.942556381225586
+INFO:__main__:vgg_loss_2: 42.191978454589844
+INFO:__main__:vgg_loss_3: 30.12472152709961
+INFO:__main__:vgg_loss_4: 52.86692810058594
+INFO:__main__:vgg_loss_5: 1.7859548330307007
+INFO:__main__:validation_loss: 1794.59765625
+INFO:__main__:global_step: 57251
+INFO:__main__:kl_loss: 233.87469482421875
+INFO:__main__:kl_weight: 0.29000070691108704
+INFO:__main__:likelihood_loss: 1617.7073974609375
+INFO:__main__:loss: 1685.53125
+INFO:__main__:vgg_gram_loss_0: 4.5518317222595215
+INFO:__main__:vgg_gram_loss_1: 15.912202835083008
+INFO:__main__:vgg_gram_loss_2: 23.057598114013672
+INFO:__main__:vgg_gram_loss_3: 18.661273956298828
+INFO:__main__:vgg_gram_loss_4: 83.39867401123047
+INFO:__main__:vgg_gram_loss_5: 0.185943141579628
+INFO:__main__:vgg_loss_0: 12.52323055267334
+INFO:__main__:vgg_loss_1: 30.470561981201172
+INFO:__main__:vgg_loss_2: 45.55154800415039
+INFO:__main__:vgg_loss_3: 32.33949279785156
+INFO:__main__:vgg_loss_4: 55.08700942993164
+INFO:__main__:vgg_loss_5: 1.8021098375320435
+INFO:__main__:validation_loss: 1676.0894775390625
+INFO:__main__:global_step: 57501
+INFO:__main__:kl_loss: 240.29698181152344
+INFO:__main__:kl_weight: 0.30000069737434387
+INFO:__main__:likelihood_loss: 1532.528076171875
+INFO:__main__:loss: 1604.6173095703125
+INFO:__main__:vgg_gram_loss_0: 6.826887607574463
+INFO:__main__:vgg_gram_loss_1: 22.4799861907959
+INFO:__main__:vgg_gram_loss_2: 23.11798095703125
+INFO:__main__:vgg_gram_loss_3: 18.424182891845703
+INFO:__main__:vgg_gram_loss_4: 73.97643280029297
+INFO:__main__:vgg_gram_loss_5: 0.1532263159751892
+INFO:__main__:vgg_loss_0: 11.317983627319336
+INFO:__main__:vgg_loss_1: 28.035123825073242
+INFO:__main__:vgg_loss_2: 40.7356071472168
+INFO:__main__:vgg_loss_3: 29.25511932373047
+INFO:__main__:vgg_loss_4: 50.573848724365234
+INFO:__main__:vgg_loss_5: 1.6092559099197388
+INFO:__main__:validation_loss: 2205.823486328125
+INFO:__main__:global_step: 57751
+INFO:__main__:kl_loss: 193.55615234375
+INFO:__main__:kl_weight: 0.3100006878376007
+INFO:__main__:likelihood_loss: 1320.489013671875
+INFO:__main__:loss: 1380.4915771484375
+INFO:__main__:vgg_gram_loss_0: 8.774993896484375
+INFO:__main__:vgg_gram_loss_1: 16.969257354736328
+INFO:__main__:vgg_gram_loss_2: 16.205549240112305
+INFO:__main__:vgg_gram_loss_3: 13.636611938476562
+INFO:__main__:vgg_gram_loss_4: 65.27970123291016
+INFO:__main__:vgg_gram_loss_5: 0.16517125070095062
+INFO:__main__:vgg_loss_0: 9.513181686401367
+INFO:__main__:vgg_loss_1: 23.318103790283203
+INFO:__main__:vgg_loss_2: 35.0458984375
+INFO:__main__:vgg_loss_3: 25.928007125854492
+INFO:__main__:vgg_loss_4: 47.58470153808594
+INFO:__main__:vgg_loss_5: 1.6766400337219238
+INFO:__main__:validation_loss: 2050.588134765625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 58001
+INFO:__main__:kl_loss: 243.0958709716797
+INFO:__main__:kl_weight: 0.32000067830085754
+INFO:__main__:likelihood_loss: 1726.4154052734375
+INFO:__main__:loss: 1804.206298828125
+INFO:__main__:vgg_gram_loss_0: 6.8467535972595215
+INFO:__main__:vgg_gram_loss_1: 20.26763343811035
+INFO:__main__:vgg_gram_loss_2: 28.784292221069336
+INFO:__main__:vgg_gram_loss_3: 20.6180477142334
+INFO:__main__:vgg_gram_loss_4: 82.578369140625
+INFO:__main__:vgg_gram_loss_5: 0.19503463804721832
+INFO:__main__:vgg_loss_0: 14.219518661499023
+INFO:__main__:vgg_loss_1: 34.867950439453125
+INFO:__main__:vgg_loss_2: 46.92289352416992
+INFO:__main__:vgg_loss_3: 32.465965270996094
+INFO:__main__:vgg_loss_4: 55.683982849121094
+INFO:__main__:vgg_loss_5: 1.8326365947723389
+INFO:__main__:validation_loss: 1689.6268310546875
+INFO:__main__:global_step: 58251
+INFO:__main__:kl_loss: 205.43701171875
+INFO:__main__:kl_weight: 0.3300006687641144
+INFO:__main__:likelihood_loss: 1348.15673828125
+INFO:__main__:loss: 1415.9510498046875
+INFO:__main__:vgg_gram_loss_0: 8.835326194763184
+INFO:__main__:vgg_gram_loss_1: 16.253568649291992
+INFO:__main__:vgg_gram_loss_2: 17.209184646606445
+INFO:__main__:vgg_gram_loss_3: 14.403388977050781
+INFO:__main__:vgg_gram_loss_4: 69.08394622802734
+INFO:__main__:vgg_gram_loss_5: 0.16471914947032928
+INFO:__main__:vgg_loss_0: 10.182188987731934
+INFO:__main__:vgg_loss_1: 23.86077880859375
+INFO:__main__:vgg_loss_2: 35.344940185546875
+INFO:__main__:vgg_loss_3: 25.976760864257812
+INFO:__main__:vgg_loss_4: 46.7400016784668
+INFO:__main__:vgg_loss_5: 1.5765422582626343
+INFO:__main__:validation_loss: 1976.2889404296875
+INFO:__main__:global_step: 58501
+INFO:__main__:kl_loss: 219.08233642578125
+INFO:__main__:kl_weight: 0.3400006592273712
+INFO:__main__:likelihood_loss: 1532.663330078125
+INFO:__main__:loss: 1607.1514892578125
+INFO:__main__:vgg_gram_loss_0: 7.922638893127441
+INFO:__main__:vgg_gram_loss_1: 15.997161865234375
+INFO:__main__:vgg_gram_loss_2: 19.515382766723633
+INFO:__main__:vgg_gram_loss_3: 15.397544860839844
+INFO:__main__:vgg_gram_loss_4: 77.22454071044922
+INFO:__main__:vgg_gram_loss_5: 0.19094879925251007
+INFO:__main__:vgg_loss_0: 11.872652053833008
+INFO:__main__:vgg_loss_1: 27.999563217163086
+INFO:__main__:vgg_loss_2: 42.17500305175781
+INFO:__main__:vgg_loss_3: 30.67888832092285
+INFO:__main__:vgg_loss_4: 55.72467803955078
+INFO:__main__:vgg_loss_5: 1.8336634635925293
+INFO:__main__:validation_loss: 1927.972412109375
+INFO:__main__:global_step: 58751
+INFO:__main__:kl_loss: 215.67520141601562
+INFO:__main__:kl_weight: 0.35000064969062805
+INFO:__main__:likelihood_loss: 1681.076416015625
+INFO:__main__:loss: 1756.5628662109375
+INFO:__main__:vgg_gram_loss_0: 5.610507488250732
+INFO:__main__:vgg_gram_loss_1: 16.57940101623535
+INFO:__main__:vgg_gram_loss_2: 22.824447631835938
+INFO:__main__:vgg_gram_loss_3: 21.823196411132812
+INFO:__main__:vgg_gram_loss_4: 88.89098358154297
+INFO:__main__:vgg_gram_loss_5: 0.19646035134792328
+INFO:__main__:vgg_loss_0: 13.272629737854004
+INFO:__main__:vgg_loss_1: 32.366817474365234
+INFO:__main__:vgg_loss_2: 45.62297058105469
+INFO:__main__:vgg_loss_3: 32.30221939086914
+INFO:__main__:vgg_loss_4: 54.84516906738281
+INFO:__main__:vgg_loss_5: 1.8804705142974854
+INFO:__main__:validation_loss: 2150.22314453125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 59001
+INFO:__main__:kl_loss: 193.33583068847656
+INFO:__main__:kl_weight: 0.3600006401538849
+INFO:__main__:likelihood_loss: 1429.91064453125
+INFO:__main__:loss: 1499.51171875
+INFO:__main__:vgg_gram_loss_0: 6.279611110687256
+INFO:__main__:vgg_gram_loss_1: 18.69470977783203
+INFO:__main__:vgg_gram_loss_2: 21.265682220458984
+INFO:__main__:vgg_gram_loss_3: 15.021896362304688
+INFO:__main__:vgg_gram_loss_4: 69.36939239501953
+INFO:__main__:vgg_gram_loss_5: 0.17039336264133453
+INFO:__main__:vgg_loss_0: 10.9222412109375
+INFO:__main__:vgg_loss_1: 26.75729751586914
+INFO:__main__:vgg_loss_2: 39.64116668701172
+INFO:__main__:vgg_loss_3: 28.392799377441406
+INFO:__main__:vgg_loss_4: 47.86614990234375
+INFO:__main__:vgg_loss_5: 1.6008058786392212
+INFO:__main__:validation_loss: 2160.015625
+INFO:__main__:global_step: 59251
+INFO:__main__:kl_loss: 180.37911987304688
+INFO:__main__:kl_weight: 0.3700006306171417
+INFO:__main__:likelihood_loss: 1618.112548828125
+INFO:__main__:loss: 1684.8529052734375
+INFO:__main__:vgg_gram_loss_0: 5.837273597717285
+INFO:__main__:vgg_gram_loss_1: 18.40519142150879
+INFO:__main__:vgg_gram_loss_2: 23.166988372802734
+INFO:__main__:vgg_gram_loss_3: 19.53774070739746
+INFO:__main__:vgg_gram_loss_4: 84.2310791015625
+INFO:__main__:vgg_gram_loss_5: 0.1909681260585785
+INFO:__main__:vgg_loss_0: 10.9801607131958
+INFO:__main__:vgg_loss_1: 29.02680778503418
+INFO:__main__:vgg_loss_2: 42.57131576538086
+INFO:__main__:vgg_loss_3: 31.413135528564453
+INFO:__main__:vgg_loss_4: 56.370323181152344
+INFO:__main__:vgg_loss_5: 1.8914778232574463
+INFO:__main__:validation_loss: 2272.75927734375
+INFO:__main__:global_step: 59501
+INFO:__main__:kl_loss: 183.94268798828125
+INFO:__main__:kl_weight: 0.38000062108039856
+INFO:__main__:likelihood_loss: 1798.91015625
+INFO:__main__:loss: 1868.8084716796875
+INFO:__main__:vgg_gram_loss_0: 7.954429626464844
+INFO:__main__:vgg_gram_loss_1: 22.87876319885254
+INFO:__main__:vgg_gram_loss_2: 27.028940200805664
+INFO:__main__:vgg_gram_loss_3: 20.88850975036621
+INFO:__main__:vgg_gram_loss_4: 86.61819458007812
+INFO:__main__:vgg_gram_loss_5: 0.19709549844264984
+INFO:__main__:vgg_loss_0: 14.981225967407227
+INFO:__main__:vgg_loss_1: 38.249427795410156
+INFO:__main__:vgg_loss_2: 49.45981979370117
+INFO:__main__:vgg_loss_3: 33.42839431762695
+INFO:__main__:vgg_loss_4: 56.236228942871094
+INFO:__main__:vgg_loss_5: 1.860987663269043
+INFO:__main__:validation_loss: 1814.781494140625
+INFO:__main__:global_step: 59751
+INFO:__main__:kl_loss: 193.98171997070312
+INFO:__main__:kl_weight: 0.3900006115436554
+INFO:__main__:likelihood_loss: 1749.99755859375
+INFO:__main__:loss: 1825.6505126953125
+INFO:__main__:vgg_gram_loss_0: 7.6700310707092285
+INFO:__main__:vgg_gram_loss_1: 22.643075942993164
+INFO:__main__:vgg_gram_loss_2: 23.319944381713867
+INFO:__main__:vgg_gram_loss_3: 21.483612060546875
+INFO:__main__:vgg_gram_loss_4: 96.3400650024414
+INFO:__main__:vgg_gram_loss_5: 0.21944260597229004
+INFO:__main__:vgg_loss_0: 11.843279838562012
+INFO:__main__:vgg_loss_1: 29.51251792907715
+INFO:__main__:vgg_loss_2: 43.358131408691406
+INFO:__main__:vgg_loss_3: 32.64407730102539
+INFO:__main__:vgg_loss_4: 58.97665786743164
+INFO:__main__:vgg_loss_5: 1.9886924028396606
+INFO:__main__:validation_loss: 1773.67041015625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 60001
+INFO:__main__:kl_loss: 197.79312133789062
+INFO:__main__:kl_weight: 0.40000060200691223
+INFO:__main__:likelihood_loss: 1480.00732421875
+INFO:__main__:loss: 1559.124755859375
+INFO:__main__:vgg_gram_loss_0: 8.505050659179688
+INFO:__main__:vgg_gram_loss_1: 16.957721710205078
+INFO:__main__:vgg_gram_loss_2: 19.72979736328125
+INFO:__main__:vgg_gram_loss_3: 15.914692878723145
+INFO:__main__:vgg_gram_loss_4: 73.58789825439453
+INFO:__main__:vgg_gram_loss_5: 0.15875029563903809
+INFO:__main__:vgg_loss_0: 12.003179550170898
+INFO:__main__:vgg_loss_1: 26.718629837036133
+INFO:__main__:vgg_loss_2: 38.910369873046875
+INFO:__main__:vgg_loss_3: 28.952842712402344
+INFO:__main__:vgg_loss_4: 52.769004821777344
+INFO:__main__:vgg_loss_5: 1.7935148477554321
+INFO:__main__:validation_loss: 2121.648681640625
+INFO:__main__:global_step: 60251
+INFO:__main__:kl_loss: 181.4539031982422
+INFO:__main__:kl_weight: 0.41000059247016907
+INFO:__main__:likelihood_loss: 1709.8804931640625
+INFO:__main__:loss: 1784.2767333984375
+INFO:__main__:vgg_gram_loss_0: 8.733294486999512
+INFO:__main__:vgg_gram_loss_1: 17.923580169677734
+INFO:__main__:vgg_gram_loss_2: 23.080129623413086
+INFO:__main__:vgg_gram_loss_3: 19.587923049926758
+INFO:__main__:vgg_gram_loss_4: 86.67509460449219
+INFO:__main__:vgg_gram_loss_5: 0.20076461136341095
+INFO:__main__:vgg_loss_0: 12.751214981079102
+INFO:__main__:vgg_loss_1: 31.6510009765625
+INFO:__main__:vgg_loss_2: 44.42028045654297
+INFO:__main__:vgg_loss_3: 33.183746337890625
+INFO:__main__:vgg_loss_4: 61.778175354003906
+INFO:__main__:vgg_loss_5: 1.9908862113952637
+INFO:__main__:validation_loss: 1997.012939453125
+INFO:__main__:global_step: 60501
+INFO:__main__:kl_loss: 170.43324279785156
+INFO:__main__:kl_weight: 0.4200005829334259
+INFO:__main__:likelihood_loss: 1556.390869140625
+INFO:__main__:loss: 1627.972900390625
+INFO:__main__:vgg_gram_loss_0: 10.932792663574219
+INFO:__main__:vgg_gram_loss_1: 22.852108001708984
+INFO:__main__:vgg_gram_loss_2: 23.604936599731445
+INFO:__main__:vgg_gram_loss_3: 18.7296199798584
+INFO:__main__:vgg_gram_loss_4: 75.07914733886719
+INFO:__main__:vgg_gram_loss_5: 0.17030248045921326
+INFO:__main__:vgg_loss_0: 10.895041465759277
+INFO:__main__:vgg_loss_1: 25.68528175354004
+INFO:__main__:vgg_loss_2: 39.50252151489258
+INFO:__main__:vgg_loss_3: 29.35719108581543
+INFO:__main__:vgg_loss_4: 52.71746063232422
+INFO:__main__:vgg_loss_5: 1.751753568649292
+INFO:__main__:validation_loss: 1990.8094482421875
+INFO:__main__:global_step: 60751
+INFO:__main__:kl_loss: 183.24961853027344
+INFO:__main__:kl_weight: 0.43000057339668274
+INFO:__main__:likelihood_loss: 2108.6884765625
+INFO:__main__:loss: 2187.48583984375
+INFO:__main__:vgg_gram_loss_0: 9.331361770629883
+INFO:__main__:vgg_gram_loss_1: 32.97922897338867
+INFO:__main__:vgg_gram_loss_2: 31.343585968017578
+INFO:__main__:vgg_gram_loss_3: 25.617319107055664
+INFO:__main__:vgg_gram_loss_4: 102.14390563964844
+INFO:__main__:vgg_gram_loss_5: 0.19482062757015228
+INFO:__main__:vgg_loss_0: 18.63156509399414
+INFO:__main__:vgg_loss_1: 41.982215881347656
+INFO:__main__:vgg_loss_2: 55.34001922607422
+INFO:__main__:vgg_loss_3: 38.46071243286133
+INFO:__main__:vgg_loss_4: 63.79774475097656
+INFO:__main__:vgg_loss_5: 1.9152261018753052
+INFO:__main__:validation_loss: 1520.69873046875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 61001
+INFO:__main__:kl_loss: 167.36868286132812
+INFO:__main__:kl_weight: 0.4400005638599396
+INFO:__main__:likelihood_loss: 1539.850830078125
+INFO:__main__:loss: 1613.4931640625
+INFO:__main__:vgg_gram_loss_0: 9.395010948181152
+INFO:__main__:vgg_gram_loss_1: 21.405475616455078
+INFO:__main__:vgg_gram_loss_2: 24.553144454956055
+INFO:__main__:vgg_gram_loss_3: 17.760648727416992
+INFO:__main__:vgg_gram_loss_4: 74.90666961669922
+INFO:__main__:vgg_gram_loss_5: 0.1572108119726181
+INFO:__main__:vgg_loss_0: 11.96183967590332
+INFO:__main__:vgg_loss_1: 29.305347442626953
+INFO:__main__:vgg_loss_2: 40.379425048828125
+INFO:__main__:vgg_loss_3: 27.906919479370117
+INFO:__main__:vgg_loss_4: 48.647972106933594
+INFO:__main__:vgg_loss_5: 1.5904710292816162
+INFO:__main__:validation_loss: 1659.5078125
+INFO:__main__:global_step: 61251
+INFO:__main__:kl_loss: 150.88949584960938
+INFO:__main__:kl_weight: 0.4500005543231964
+INFO:__main__:likelihood_loss: 1341.9801025390625
+INFO:__main__:loss: 1409.8804931640625
+INFO:__main__:vgg_gram_loss_0: 7.739903926849365
+INFO:__main__:vgg_gram_loss_1: 17.0947208404541
+INFO:__main__:vgg_gram_loss_2: 18.212779998779297
+INFO:__main__:vgg_gram_loss_3: 14.0092134475708
+INFO:__main__:vgg_gram_loss_4: 65.4217758178711
+INFO:__main__:vgg_gram_loss_5: 0.1537393480539322
+INFO:__main__:vgg_loss_0: 9.86634349822998
+INFO:__main__:vgg_loss_1: 24.50064468383789
+INFO:__main__:vgg_loss_2: 36.40787124633789
+INFO:__main__:vgg_loss_3: 26.246822357177734
+INFO:__main__:vgg_loss_4: 47.179168701171875
+INFO:__main__:vgg_loss_5: 1.5630505084991455
+INFO:__main__:validation_loss: 2023.6031494140625
+INFO:__main__:global_step: 61501
+INFO:__main__:kl_loss: 172.0615234375
+INFO:__main__:kl_weight: 0.46000054478645325
+INFO:__main__:likelihood_loss: 1423.3935546875
+INFO:__main__:loss: 1502.5419921875
+INFO:__main__:vgg_gram_loss_0: 6.877858638763428
+INFO:__main__:vgg_gram_loss_1: 18.043676376342773
+INFO:__main__:vgg_gram_loss_2: 20.73009490966797
+INFO:__main__:vgg_gram_loss_3: 17.797643661499023
+INFO:__main__:vgg_gram_loss_4: 74.64494323730469
+INFO:__main__:vgg_gram_loss_5: 0.1636204570531845
+INFO:__main__:vgg_loss_0: 9.85766315460205
+INFO:__main__:vgg_loss_1: 23.75548553466797
+INFO:__main__:vgg_loss_2: 35.95502853393555
+INFO:__main__:vgg_loss_3: 27.236677169799805
+INFO:__main__:vgg_loss_4: 48.01709747314453
+INFO:__main__:vgg_loss_5: 1.598944067955017
+INFO:__main__:validation_loss: 2065.132080078125
+INFO:__main__:global_step: 61751
+INFO:__main__:kl_loss: 153.73365783691406
+INFO:__main__:kl_weight: 0.4700005352497101
+INFO:__main__:likelihood_loss: 1526.9849853515625
+INFO:__main__:loss: 1599.2398681640625
+INFO:__main__:vgg_gram_loss_0: 7.539271831512451
+INFO:__main__:vgg_gram_loss_1: 17.918149948120117
+INFO:__main__:vgg_gram_loss_2: 24.20075798034668
+INFO:__main__:vgg_gram_loss_3: 18.901962280273438
+INFO:__main__:vgg_gram_loss_4: 77.95181274414062
+INFO:__main__:vgg_gram_loss_5: 0.16425113379955292
+INFO:__main__:vgg_loss_0: 11.178567886352539
+INFO:__main__:vgg_loss_1: 28.170799255371094
+INFO:__main__:vgg_loss_2: 40.80046463012695
+INFO:__main__:vgg_loss_3: 28.324134826660156
+INFO:__main__:vgg_loss_4: 48.68288040161133
+INFO:__main__:vgg_loss_5: 1.5639485120773315
+INFO:__main__:validation_loss: 1952.921142578125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 62001
+INFO:__main__:kl_loss: 157.3220977783203
+INFO:__main__:kl_weight: 0.4800005257129669
+INFO:__main__:likelihood_loss: 1395.3570556640625
+INFO:__main__:loss: 1470.8717041015625
+INFO:__main__:vgg_gram_loss_0: 8.790541648864746
+INFO:__main__:vgg_gram_loss_1: 16.917423248291016
+INFO:__main__:vgg_gram_loss_2: 19.832860946655273
+INFO:__main__:vgg_gram_loss_3: 15.707832336425781
+INFO:__main__:vgg_gram_loss_4: 69.8316650390625
+INFO:__main__:vgg_gram_loss_5: 0.15188977122306824
+INFO:__main__:vgg_loss_0: 10.436554908752441
+INFO:__main__:vgg_loss_1: 25.277132034301758
+INFO:__main__:vgg_loss_2: 36.39839172363281
+INFO:__main__:vgg_loss_3: 26.555042266845703
+INFO:__main__:vgg_loss_4: 47.579185485839844
+INFO:__main__:vgg_loss_5: 1.5928845405578613
+INFO:__main__:validation_loss: 1986.4376220703125
+INFO:__main__:global_step: 62251
+INFO:__main__:kl_loss: 155.29299926757812
+INFO:__main__:kl_weight: 0.49000051617622375
+INFO:__main__:likelihood_loss: 1563.59130859375
+INFO:__main__:loss: 1639.6849365234375
+INFO:__main__:vgg_gram_loss_0: 5.029366970062256
+INFO:__main__:vgg_gram_loss_1: 19.014877319335938
+INFO:__main__:vgg_gram_loss_2: 24.09842872619629
+INFO:__main__:vgg_gram_loss_3: 17.659481048583984
+INFO:__main__:vgg_gram_loss_4: 78.6868896484375
+INFO:__main__:vgg_gram_loss_5: 0.18552082777023315
+INFO:__main__:vgg_loss_0: 11.27612018585205
+INFO:__main__:vgg_loss_1: 29.093385696411133
+INFO:__main__:vgg_loss_2: 42.61553192138672
+INFO:__main__:vgg_loss_3: 30.585060119628906
+INFO:__main__:vgg_loss_4: 52.72942352294922
+INFO:__main__:vgg_loss_5: 1.7441712617874146
+INFO:__main__:validation_loss: 1859.8572998046875
+INFO:__main__:global_step: 62501
+INFO:__main__:kl_loss: 181.54559326171875
+INFO:__main__:kl_weight: 0.5000004768371582
+INFO:__main__:likelihood_loss: 1895.2548828125
+INFO:__main__:loss: 1986.0277099609375
+INFO:__main__:vgg_gram_loss_0: 7.939996242523193
+INFO:__main__:vgg_gram_loss_1: 19.90495491027832
+INFO:__main__:vgg_gram_loss_2: 26.79852294921875
+INFO:__main__:vgg_gram_loss_3: 21.490812301635742
+INFO:__main__:vgg_gram_loss_4: 96.40921783447266
+INFO:__main__:vgg_gram_loss_5: 0.2147810310125351
+INFO:__main__:vgg_loss_0: 14.152655601501465
+INFO:__main__:vgg_loss_1: 34.47138595581055
+INFO:__main__:vgg_loss_2: 52.4765625
+INFO:__main__:vgg_loss_3: 37.76268768310547
+INFO:__main__:vgg_loss_4: 65.34140014648438
+INFO:__main__:vgg_loss_5: 2.0880179405212402
+INFO:__main__:validation_loss: 2046.990478515625
+INFO:__main__:global_step: 62751
+INFO:__main__:kl_loss: 154.36138916015625
+INFO:__main__:kl_weight: 0.510000467300415
+INFO:__main__:likelihood_loss: 1456.501220703125
+INFO:__main__:loss: 1535.2255859375
+INFO:__main__:vgg_gram_loss_0: 5.498467445373535
+INFO:__main__:vgg_gram_loss_1: 17.481721878051758
+INFO:__main__:vgg_gram_loss_2: 20.46467399597168
+INFO:__main__:vgg_gram_loss_3: 17.913406372070312
+INFO:__main__:vgg_gram_loss_4: 73.32341003417969
+INFO:__main__:vgg_gram_loss_5: 0.14709335565567017
+INFO:__main__:vgg_loss_0: 12.278597831726074
+INFO:__main__:vgg_loss_1: 29.25261116027832
+INFO:__main__:vgg_loss_2: 38.8270263671875
+INFO:__main__:vgg_loss_3: 27.46396255493164
+INFO:__main__:vgg_loss_4: 47.117122650146484
+INFO:__main__:vgg_loss_5: 1.532151460647583
+INFO:__main__:validation_loss: 1875.308837890625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 63001
+INFO:__main__:kl_loss: 158.30812072753906
+INFO:__main__:kl_weight: 0.5200004577636719
+INFO:__main__:likelihood_loss: 1332.3291015625
+INFO:__main__:loss: 1414.6494140625
+INFO:__main__:vgg_gram_loss_0: 6.255046844482422
+INFO:__main__:vgg_gram_loss_1: 14.249056816101074
+INFO:__main__:vgg_gram_loss_2: 16.584993362426758
+INFO:__main__:vgg_gram_loss_3: 13.93603801727295
+INFO:__main__:vgg_gram_loss_4: 71.7183609008789
+INFO:__main__:vgg_gram_loss_5: 0.15620115399360657
+INFO:__main__:vgg_loss_0: 9.447062492370605
+INFO:__main__:vgg_loss_1: 24.19516372680664
+INFO:__main__:vgg_loss_2: 35.44914627075195
+INFO:__main__:vgg_loss_3: 25.659456253051758
+INFO:__main__:vgg_loss_4: 47.22391891479492
+INFO:__main__:vgg_loss_5: 1.5913810729980469
+INFO:__main__:validation_loss: 1799.3363037109375
+INFO:__main__:global_step: 63251
+INFO:__main__:kl_loss: 148.7079315185547
+INFO:__main__:kl_weight: 0.5300004482269287
+INFO:__main__:likelihood_loss: 1577.3358154296875
+INFO:__main__:loss: 1656.151123046875
+INFO:__main__:vgg_gram_loss_0: 7.008537292480469
+INFO:__main__:vgg_gram_loss_1: 19.99354362487793
+INFO:__main__:vgg_gram_loss_2: 21.231698989868164
+INFO:__main__:vgg_gram_loss_3: 17.07937240600586
+INFO:__main__:vgg_gram_loss_4: 80.08322143554688
+INFO:__main__:vgg_gram_loss_5: 0.183013454079628
+INFO:__main__:vgg_loss_0: 12.919233322143555
+INFO:__main__:vgg_loss_1: 29.230680465698242
+INFO:__main__:vgg_loss_2: 41.317047119140625
+INFO:__main__:vgg_loss_3: 30.217702865600586
+INFO:__main__:vgg_loss_4: 54.374755859375
+INFO:__main__:vgg_loss_5: 1.8283731937408447
+INFO:__main__:validation_loss: 1956.83349609375
+INFO:__main__:global_step: 63501
+INFO:__main__:kl_loss: 148.82266235351562
+INFO:__main__:kl_weight: 0.5400004386901855
+INFO:__main__:likelihood_loss: 1506.35009765625
+INFO:__main__:loss: 1586.71435546875
+INFO:__main__:vgg_gram_loss_0: 11.792441368103027
+INFO:__main__:vgg_gram_loss_1: 22.938583374023438
+INFO:__main__:vgg_gram_loss_2: 22.563390731811523
+INFO:__main__:vgg_gram_loss_3: 17.500226974487305
+INFO:__main__:vgg_gram_loss_4: 72.64840698242188
+INFO:__main__:vgg_gram_loss_5: 0.1615828573703766
+INFO:__main__:vgg_loss_0: 10.978843688964844
+INFO:__main__:vgg_loss_1: 26.082365036010742
+INFO:__main__:vgg_loss_2: 38.441402435302734
+INFO:__main__:vgg_loss_3: 28.088895797729492
+INFO:__main__:vgg_loss_4: 48.461326599121094
+INFO:__main__:vgg_loss_5: 1.6125644445419312
+INFO:__main__:validation_loss: 1992.5570068359375
+INFO:__main__:global_step: 63751
+INFO:__main__:kl_loss: 144.92318725585938
+INFO:__main__:kl_weight: 0.5500004291534424
+INFO:__main__:likelihood_loss: 1538.49755859375
+INFO:__main__:loss: 1618.205322265625
+INFO:__main__:vgg_gram_loss_0: 9.499394416809082
+INFO:__main__:vgg_gram_loss_1: 17.774600982666016
+INFO:__main__:vgg_gram_loss_2: 20.655994415283203
+INFO:__main__:vgg_gram_loss_3: 16.993106842041016
+INFO:__main__:vgg_gram_loss_4: 77.11156463623047
+INFO:__main__:vgg_gram_loss_5: 0.17355114221572876
+INFO:__main__:vgg_loss_0: 11.731160163879395
+INFO:__main__:vgg_loss_1: 27.712064743041992
+INFO:__main__:vgg_loss_2: 41.247802734375
+INFO:__main__:vgg_loss_3: 30.156343460083008
+INFO:__main__:vgg_loss_4: 52.89293670654297
+INFO:__main__:vgg_loss_5: 1.7510243654251099
+INFO:__main__:validation_loss: 1677.9342041015625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 64001
+INFO:__main__:kl_loss: 130.691650390625
+INFO:__main__:kl_weight: 0.5600004196166992
+INFO:__main__:likelihood_loss: 1296.4849853515625
+INFO:__main__:loss: 1369.67236328125
+INFO:__main__:vgg_gram_loss_0: 4.836056709289551
+INFO:__main__:vgg_gram_loss_1: 15.333206176757812
+INFO:__main__:vgg_gram_loss_2: 18.758073806762695
+INFO:__main__:vgg_gram_loss_3: 14.820637702941895
+INFO:__main__:vgg_gram_loss_4: 63.64361572265625
+INFO:__main__:vgg_gram_loss_5: 0.1568421572446823
+INFO:__main__:vgg_loss_0: 9.76473617553711
+INFO:__main__:vgg_loss_1: 23.370521545410156
+INFO:__main__:vgg_loss_2: 35.16050720214844
+INFO:__main__:vgg_loss_3: 25.989593505859375
+INFO:__main__:vgg_loss_4: 45.89315414428711
+INFO:__main__:vgg_loss_5: 1.570067286491394
+INFO:__main__:validation_loss: 1913.1480712890625
+INFO:__main__:global_step: 64251
+INFO:__main__:kl_loss: 131.55601501464844
+INFO:__main__:kl_weight: 0.570000410079956
+INFO:__main__:likelihood_loss: 1356.806396484375
+INFO:__main__:loss: 1431.7933349609375
+INFO:__main__:vgg_gram_loss_0: 5.879093170166016
+INFO:__main__:vgg_gram_loss_1: 12.707127571105957
+INFO:__main__:vgg_gram_loss_2: 16.545490264892578
+INFO:__main__:vgg_gram_loss_3: 14.249430656433105
+INFO:__main__:vgg_gram_loss_4: 71.69304656982422
+INFO:__main__:vgg_gram_loss_5: 0.17117194831371307
+INFO:__main__:vgg_loss_0: 10.482495307922363
+INFO:__main__:vgg_loss_1: 25.258399963378906
+INFO:__main__:vgg_loss_2: 37.025882720947266
+INFO:__main__:vgg_loss_3: 27.014118194580078
+INFO:__main__:vgg_loss_4: 48.70873260498047
+INFO:__main__:vgg_loss_5: 1.6262907981872559
+INFO:__main__:validation_loss: 1797.1842041015625
+INFO:__main__:global_step: 64501
+INFO:__main__:kl_loss: 142.2903289794922
+INFO:__main__:kl_weight: 0.5800004005432129
+INFO:__main__:likelihood_loss: 1376.1630859375
+INFO:__main__:loss: 1458.6915283203125
+INFO:__main__:vgg_gram_loss_0: 4.241140842437744
+INFO:__main__:vgg_gram_loss_1: 17.07489585876465
+INFO:__main__:vgg_gram_loss_2: 20.38210105895996
+INFO:__main__:vgg_gram_loss_3: 16.10785484313965
+INFO:__main__:vgg_gram_loss_4: 68.23886108398438
+INFO:__main__:vgg_gram_loss_5: 0.16461364924907684
+INFO:__main__:vgg_loss_0: 9.85578727722168
+INFO:__main__:vgg_loss_1: 24.325496673583984
+INFO:__main__:vgg_loss_2: 36.98074722290039
+INFO:__main__:vgg_loss_3: 27.342815399169922
+INFO:__main__:vgg_loss_4: 48.85261154174805
+INFO:__main__:vgg_loss_5: 1.6656838655471802
+INFO:__main__:validation_loss: 1885.7972412109375
+INFO:__main__:global_step: 64751
+INFO:__main__:kl_loss: 149.5614471435547
+INFO:__main__:kl_weight: 0.5900003910064697
+INFO:__main__:likelihood_loss: 1452.321533203125
+INFO:__main__:loss: 1540.5628662109375
+INFO:__main__:vgg_gram_loss_0: 9.444628715515137
+INFO:__main__:vgg_gram_loss_1: 17.266794204711914
+INFO:__main__:vgg_gram_loss_2: 20.748369216918945
+INFO:__main__:vgg_gram_loss_3: 15.716171264648438
+INFO:__main__:vgg_gram_loss_4: 73.0754623413086
+INFO:__main__:vgg_gram_loss_5: 0.16180644929409027
+INFO:__main__:vgg_loss_0: 10.525382041931152
+INFO:__main__:vgg_loss_1: 25.975059509277344
+INFO:__main__:vgg_loss_2: 38.856319427490234
+INFO:__main__:vgg_loss_3: 27.922163009643555
+INFO:__main__:vgg_loss_4: 49.148658752441406
+INFO:__main__:vgg_loss_5: 1.6234883069992065
+INFO:__main__:validation_loss: 1734.2523193359375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 65001
+INFO:__main__:kl_loss: 130.78785705566406
+INFO:__main__:kl_weight: 0.6000003814697266
+INFO:__main__:likelihood_loss: 1521.681396484375
+INFO:__main__:loss: 1600.1541748046875
+INFO:__main__:vgg_gram_loss_0: 5.774623394012451
+INFO:__main__:vgg_gram_loss_1: 15.3505277633667
+INFO:__main__:vgg_gram_loss_2: 22.47835350036621
+INFO:__main__:vgg_gram_loss_3: 17.383214950561523
+INFO:__main__:vgg_gram_loss_4: 76.5810775756836
+INFO:__main__:vgg_gram_loss_5: 0.16186819970607758
+INFO:__main__:vgg_loss_0: 11.943792343139648
+INFO:__main__:vgg_loss_1: 28.103734970092773
+INFO:__main__:vgg_loss_2: 41.24433898925781
+INFO:__main__:vgg_loss_3: 29.874759674072266
+INFO:__main__:vgg_loss_4: 53.72663116455078
+INFO:__main__:vgg_loss_5: 1.7133642435073853
+INFO:__main__:validation_loss: 1988.6507568359375
+INFO:__main__:global_step: 65251
+INFO:__main__:kl_loss: 133.06964111328125
+INFO:__main__:kl_weight: 0.6100003719329834
+INFO:__main__:likelihood_loss: 1607.8204345703125
+INFO:__main__:loss: 1688.992919921875
+INFO:__main__:vgg_gram_loss_0: 7.787967205047607
+INFO:__main__:vgg_gram_loss_1: 18.953765869140625
+INFO:__main__:vgg_gram_loss_2: 24.768461227416992
+INFO:__main__:vgg_gram_loss_3: 19.00188446044922
+INFO:__main__:vgg_gram_loss_4: 77.5338363647461
+INFO:__main__:vgg_gram_loss_5: 0.17573629319667816
+INFO:__main__:vgg_loss_0: 13.058457374572754
+INFO:__main__:vgg_loss_1: 32.80005645751953
+INFO:__main__:vgg_loss_2: 43.70207214355469
+INFO:__main__:vgg_loss_3: 30.06410026550293
+INFO:__main__:vgg_loss_4: 51.99100875854492
+INFO:__main__:vgg_loss_5: 1.7267571687698364
+INFO:__main__:validation_loss: 2065.5517578125
+INFO:__main__:global_step: 65501
+INFO:__main__:kl_loss: 137.739501953125
+INFO:__main__:kl_weight: 0.6200003623962402
+INFO:__main__:likelihood_loss: 1674.1041259765625
+INFO:__main__:loss: 1759.502685546875
+INFO:__main__:vgg_gram_loss_0: 7.077807903289795
+INFO:__main__:vgg_gram_loss_1: 17.701295852661133
+INFO:__main__:vgg_gram_loss_2: 21.997289657592773
+INFO:__main__:vgg_gram_loss_3: 19.126741409301758
+INFO:__main__:vgg_gram_loss_4: 86.34031677246094
+INFO:__main__:vgg_gram_loss_5: 0.19935135543346405
+INFO:__main__:vgg_loss_0: 12.725826263427734
+INFO:__main__:vgg_loss_1: 29.975820541381836
+INFO:__main__:vgg_loss_2: 44.98339080810547
+INFO:__main__:vgg_loss_3: 33.735801696777344
+INFO:__main__:vgg_loss_4: 59.039859771728516
+INFO:__main__:vgg_loss_5: 1.9173345565795898
+INFO:__main__:validation_loss: 1923.3988037109375
+INFO:__main__:global_step: 65751
+INFO:__main__:kl_loss: 143.68910217285156
+INFO:__main__:kl_weight: 0.6300003528594971
+INFO:__main__:likelihood_loss: 1703.1085205078125
+INFO:__main__:loss: 1793.6326904296875
+INFO:__main__:vgg_gram_loss_0: 4.728102207183838
+INFO:__main__:vgg_gram_loss_1: 21.864030838012695
+INFO:__main__:vgg_gram_loss_2: 25.741498947143555
+INFO:__main__:vgg_gram_loss_3: 20.477821350097656
+INFO:__main__:vgg_gram_loss_4: 85.23058319091797
+INFO:__main__:vgg_gram_loss_5: 0.18515996634960175
+INFO:__main__:vgg_loss_0: 12.521238327026367
+INFO:__main__:vgg_loss_1: 30.961219787597656
+INFO:__main__:vgg_loss_2: 47.269073486328125
+INFO:__main__:vgg_loss_3: 33.10855484008789
+INFO:__main__:vgg_loss_4: 56.687747955322266
+INFO:__main__:vgg_loss_5: 1.8466848134994507
+INFO:__main__:validation_loss: 2012.11572265625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 66001
+INFO:__main__:kl_loss: 140.4301300048828
+INFO:__main__:kl_weight: 0.6400003433227539
+INFO:__main__:likelihood_loss: 1508.233154296875
+INFO:__main__:loss: 1598.1085205078125
+INFO:__main__:vgg_gram_loss_0: 6.333065032958984
+INFO:__main__:vgg_gram_loss_1: 15.689684867858887
+INFO:__main__:vgg_gram_loss_2: 18.358047485351562
+INFO:__main__:vgg_gram_loss_3: 16.155691146850586
+INFO:__main__:vgg_gram_loss_4: 77.42984008789062
+INFO:__main__:vgg_gram_loss_5: 0.17993436753749847
+INFO:__main__:vgg_loss_0: 11.451391220092773
+INFO:__main__:vgg_loss_1: 29.1550350189209
+INFO:__main__:vgg_loss_2: 42.23038864135742
+INFO:__main__:vgg_loss_3: 29.86561393737793
+INFO:__main__:vgg_loss_4: 53.03596496582031
+INFO:__main__:vgg_loss_5: 1.761966347694397
+INFO:__main__:validation_loss: 2053.955810546875
+INFO:__main__:global_step: 66251
+INFO:__main__:kl_loss: 138.91708374023438
+INFO:__main__:kl_weight: 0.6500003337860107
+INFO:__main__:likelihood_loss: 1539.1015625
+INFO:__main__:loss: 1629.397705078125
+INFO:__main__:vgg_gram_loss_0: 8.766200065612793
+INFO:__main__:vgg_gram_loss_1: 17.94972801208496
+INFO:__main__:vgg_gram_loss_2: 23.18792724609375
+INFO:__main__:vgg_gram_loss_3: 18.552114486694336
+INFO:__main__:vgg_gram_loss_4: 78.57548522949219
+INFO:__main__:vgg_gram_loss_5: 0.16626155376434326
+INFO:__main__:vgg_loss_0: 10.935999870300293
+INFO:__main__:vgg_loss_1: 26.230270385742188
+INFO:__main__:vgg_loss_2: 40.5624885559082
+INFO:__main__:vgg_loss_3: 29.772716522216797
+INFO:__main__:vgg_loss_4: 51.44537353515625
+INFO:__main__:vgg_loss_5: 1.6757436990737915
+INFO:__main__:validation_loss: 2003.36767578125
+INFO:__main__:global_step: 66501
+INFO:__main__:kl_loss: 132.59556579589844
+INFO:__main__:kl_weight: 0.6600003242492676
+INFO:__main__:likelihood_loss: 1206.0162353515625
+INFO:__main__:loss: 1293.529296875
+INFO:__main__:vgg_gram_loss_0: 4.618630409240723
+INFO:__main__:vgg_gram_loss_1: 12.44128131866455
+INFO:__main__:vgg_gram_loss_2: 15.721549987792969
+INFO:__main__:vgg_gram_loss_3: 12.134242057800293
+INFO:__main__:vgg_gram_loss_4: 60.20072555541992
+INFO:__main__:vgg_gram_loss_5: 0.13847239315509796
+INFO:__main__:vgg_loss_0: 8.987995147705078
+INFO:__main__:vgg_loss_1: 22.49643898010254
+INFO:__main__:vgg_loss_2: 34.06087112426758
+INFO:__main__:vgg_loss_3: 24.920372009277344
+INFO:__main__:vgg_loss_4: 44.036216735839844
+INFO:__main__:vgg_loss_5: 1.4464504718780518
+INFO:__main__:validation_loss: 1772.6448974609375
+INFO:__main__:global_step: 66751
+INFO:__main__:kl_loss: 121.65013122558594
+INFO:__main__:kl_weight: 0.6700003147125244
+INFO:__main__:likelihood_loss: 1269.3359375
+INFO:__main__:loss: 1350.841552734375
+INFO:__main__:vgg_gram_loss_0: 5.661892890930176
+INFO:__main__:vgg_gram_loss_1: 14.447422981262207
+INFO:__main__:vgg_gram_loss_2: 18.51140594482422
+INFO:__main__:vgg_gram_loss_3: 13.463394165039062
+INFO:__main__:vgg_gram_loss_4: 64.50457000732422
+INFO:__main__:vgg_gram_loss_5: 0.1402585804462433
+INFO:__main__:vgg_loss_0: 9.957453727722168
+INFO:__main__:vgg_loss_1: 23.48259162902832
+INFO:__main__:vgg_loss_2: 34.38383865356445
+INFO:__main__:vgg_loss_3: 24.68840789794922
+INFO:__main__:vgg_loss_4: 43.223114013671875
+INFO:__main__:vgg_loss_5: 1.4028245210647583
+INFO:__main__:validation_loss: 1835.481689453125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 67001
+INFO:__main__:kl_loss: 139.10964965820312
+INFO:__main__:kl_weight: 0.6800003051757812
+INFO:__main__:likelihood_loss: 1433.261962890625
+INFO:__main__:loss: 1527.8565673828125
+INFO:__main__:vgg_gram_loss_0: 5.023831367492676
+INFO:__main__:vgg_gram_loss_1: 14.9660062789917
+INFO:__main__:vgg_gram_loss_2: 21.340497970581055
+INFO:__main__:vgg_gram_loss_3: 16.20815658569336
+INFO:__main__:vgg_gram_loss_4: 74.86559295654297
+INFO:__main__:vgg_gram_loss_5: 0.17463155090808868
+INFO:__main__:vgg_loss_0: 10.094112396240234
+INFO:__main__:vgg_loss_1: 25.098695755004883
+INFO:__main__:vgg_loss_2: 38.61017990112305
+INFO:__main__:vgg_loss_3: 28.326242446899414
+INFO:__main__:vgg_loss_4: 50.30464553833008
+INFO:__main__:vgg_loss_5: 1.6397920846939087
+INFO:__main__:validation_loss: 2387.88916015625
+INFO:__main__:global_step: 67251
+INFO:__main__:kl_loss: 122.5555648803711
+INFO:__main__:kl_weight: 0.6900002956390381
+INFO:__main__:likelihood_loss: 1456.716064453125
+INFO:__main__:loss: 1541.2794189453125
+INFO:__main__:vgg_gram_loss_0: 7.547300815582275
+INFO:__main__:vgg_gram_loss_1: 17.3953914642334
+INFO:__main__:vgg_gram_loss_2: 18.41188621520996
+INFO:__main__:vgg_gram_loss_3: 16.024436950683594
+INFO:__main__:vgg_gram_loss_4: 72.1866455078125
+INFO:__main__:vgg_gram_loss_5: 0.16891400516033173
+INFO:__main__:vgg_loss_0: 11.940223693847656
+INFO:__main__:vgg_loss_1: 28.571102142333984
+INFO:__main__:vgg_loss_2: 39.23056411743164
+INFO:__main__:vgg_loss_3: 28.315532684326172
+INFO:__main__:vgg_loss_4: 49.885337829589844
+INFO:__main__:vgg_loss_5: 1.6658669710159302
+INFO:__main__:validation_loss: 1848.583251953125
+INFO:__main__:global_step: 67501
+INFO:__main__:kl_loss: 122.26922607421875
+INFO:__main__:kl_weight: 0.7000002861022949
+INFO:__main__:likelihood_loss: 1392.700439453125
+INFO:__main__:loss: 1478.2889404296875
+INFO:__main__:vgg_gram_loss_0: 6.852509498596191
+INFO:__main__:vgg_gram_loss_1: 17.724885940551758
+INFO:__main__:vgg_gram_loss_2: 19.158313751220703
+INFO:__main__:vgg_gram_loss_3: 15.150555610656738
+INFO:__main__:vgg_gram_loss_4: 69.83219909667969
+INFO:__main__:vgg_gram_loss_5: 0.15133295953273773
+INFO:__main__:vgg_loss_0: 10.726128578186035
+INFO:__main__:vgg_loss_1: 26.22296905517578
+INFO:__main__:vgg_loss_2: 36.78580856323242
+INFO:__main__:vgg_loss_3: 26.704618453979492
+INFO:__main__:vgg_loss_4: 47.64253234863281
+INFO:__main__:vgg_loss_5: 1.5882437229156494
+INFO:__main__:validation_loss: 2041.9786376953125
+INFO:__main__:global_step: 67751
+INFO:__main__:kl_loss: 123.15654754638672
+INFO:__main__:kl_weight: 0.7100002765655518
+INFO:__main__:likelihood_loss: 1531.4691162109375
+INFO:__main__:loss: 1618.9102783203125
+INFO:__main__:vgg_gram_loss_0: 3.985860824584961
+INFO:__main__:vgg_gram_loss_1: 15.77622127532959
+INFO:__main__:vgg_gram_loss_2: 19.705886840820312
+INFO:__main__:vgg_gram_loss_3: 17.437889099121094
+INFO:__main__:vgg_gram_loss_4: 74.76033020019531
+INFO:__main__:vgg_gram_loss_5: 0.1705358475446701
+INFO:__main__:vgg_loss_0: 12.997247695922852
+INFO:__main__:vgg_loss_1: 31.911155700683594
+INFO:__main__:vgg_loss_2: 42.923057556152344
+INFO:__main__:vgg_loss_3: 30.909244537353516
+INFO:__main__:vgg_loss_4: 53.91752624511719
+INFO:__main__:vgg_loss_5: 1.7988643646240234
+INFO:__main__:validation_loss: 1930.9964599609375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 68001
+INFO:__main__:kl_loss: 120.66307067871094
+INFO:__main__:kl_weight: 0.7200002670288086
+INFO:__main__:likelihood_loss: 1660.11083984375
+INFO:__main__:loss: 1746.98828125
+INFO:__main__:vgg_gram_loss_0: 4.675761699676514
+INFO:__main__:vgg_gram_loss_1: 17.88255500793457
+INFO:__main__:vgg_gram_loss_2: 27.682275772094727
+INFO:__main__:vgg_gram_loss_3: 20.549049377441406
+INFO:__main__:vgg_gram_loss_4: 85.3249740600586
+INFO:__main__:vgg_gram_loss_5: 0.19199660420417786
+INFO:__main__:vgg_loss_0: 12.904491424560547
+INFO:__main__:vgg_loss_1: 29.32299041748047
+INFO:__main__:vgg_loss_2: 44.046905517578125
+INFO:__main__:vgg_loss_3: 31.680994033813477
+INFO:__main__:vgg_loss_4: 55.882442474365234
+INFO:__main__:vgg_loss_5: 1.8777046203613281
+INFO:__main__:validation_loss: 1966.3978271484375
+INFO:__main__:global_step: 68251
+INFO:__main__:kl_loss: 124.3830795288086
+INFO:__main__:kl_weight: 0.7300002574920654
+INFO:__main__:likelihood_loss: 1543.2431640625
+INFO:__main__:loss: 1634.0428466796875
+INFO:__main__:vgg_gram_loss_0: 4.517995357513428
+INFO:__main__:vgg_gram_loss_1: 17.603374481201172
+INFO:__main__:vgg_gram_loss_2: 21.23562240600586
+INFO:__main__:vgg_gram_loss_3: 16.44476890563965
+INFO:__main__:vgg_gram_loss_4: 75.24378204345703
+INFO:__main__:vgg_gram_loss_5: 0.18510901927947998
+INFO:__main__:vgg_loss_0: 12.260743141174316
+INFO:__main__:vgg_loss_1: 31.31107521057129
+INFO:__main__:vgg_loss_2: 43.945106506347656
+INFO:__main__:vgg_loss_3: 30.964588165283203
+INFO:__main__:vgg_loss_4: 53.162113189697266
+INFO:__main__:vgg_loss_5: 1.7743662595748901
+INFO:__main__:validation_loss: 1975.024658203125
+INFO:__main__:global_step: 68501
+INFO:__main__:kl_loss: 120.29890441894531
+INFO:__main__:kl_weight: 0.7400002479553223
+INFO:__main__:likelihood_loss: 1573.298095703125
+INFO:__main__:loss: 1662.3193359375
+INFO:__main__:vgg_gram_loss_0: 6.859035015106201
+INFO:__main__:vgg_gram_loss_1: 19.82688331604004
+INFO:__main__:vgg_gram_loss_2: 22.367755889892578
+INFO:__main__:vgg_gram_loss_3: 17.733129501342773
+INFO:__main__:vgg_gram_loss_4: 77.9203109741211
+INFO:__main__:vgg_gram_loss_5: 0.16752628982067108
+INFO:__main__:vgg_loss_0: 13.409920692443848
+INFO:__main__:vgg_loss_1: 30.256275177001953
+INFO:__main__:vgg_loss_2: 41.42337417602539
+INFO:__main__:vgg_loss_3: 29.832212448120117
+INFO:__main__:vgg_loss_4: 53.11823654174805
+INFO:__main__:vgg_loss_5: 1.7449506521224976
+INFO:__main__:validation_loss: 2240.46875
+INFO:__main__:global_step: 68751
+INFO:__main__:kl_loss: 125.92501068115234
+INFO:__main__:kl_weight: 0.7500002384185791
+INFO:__main__:likelihood_loss: 1751.3173828125
+INFO:__main__:loss: 1845.76123046875
+INFO:__main__:vgg_gram_loss_0: 2.8916208744049072
+INFO:__main__:vgg_gram_loss_1: 20.429540634155273
+INFO:__main__:vgg_gram_loss_2: 26.835046768188477
+INFO:__main__:vgg_gram_loss_3: 21.94961929321289
+INFO:__main__:vgg_gram_loss_4: 85.21687316894531
+INFO:__main__:vgg_gram_loss_5: 0.18313011527061462
+INFO:__main__:vgg_loss_0: 14.87549877166748
+INFO:__main__:vgg_loss_1: 37.31819534301758
+INFO:__main__:vgg_loss_2: 48.62551498413086
+INFO:__main__:vgg_loss_3: 33.33528137207031
+INFO:__main__:vgg_loss_4: 56.715576171875
+INFO:__main__:vgg_loss_5: 1.8875964879989624
+INFO:__main__:validation_loss: 2493.322998046875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 69001
+INFO:__main__:kl_loss: 125.6292724609375
+INFO:__main__:kl_weight: 0.7600002288818359
+INFO:__main__:likelihood_loss: 1536.2408447265625
+INFO:__main__:loss: 1631.7191162109375
+INFO:__main__:vgg_gram_loss_0: 5.972162246704102
+INFO:__main__:vgg_gram_loss_1: 16.523221969604492
+INFO:__main__:vgg_gram_loss_2: 21.487659454345703
+INFO:__main__:vgg_gram_loss_3: 19.820798873901367
+INFO:__main__:vgg_gram_loss_4: 83.37060546875
+INFO:__main__:vgg_gram_loss_5: 0.17578040063381195
+INFO:__main__:vgg_loss_0: 10.990117073059082
+INFO:__main__:vgg_loss_1: 27.160594940185547
+INFO:__main__:vgg_loss_2: 38.87975311279297
+INFO:__main__:vgg_loss_3: 29.251968383789062
+INFO:__main__:vgg_loss_4: 51.939918518066406
+INFO:__main__:vgg_loss_5: 1.6756130456924438
+INFO:__main__:validation_loss: 2174.975341796875
+INFO:__main__:global_step: 69251
+INFO:__main__:kl_loss: 123.2645263671875
+INFO:__main__:kl_weight: 0.7700002193450928
+INFO:__main__:likelihood_loss: 1596.554931640625
+INFO:__main__:loss: 1691.4686279296875
+INFO:__main__:vgg_gram_loss_0: 5.541776180267334
+INFO:__main__:vgg_gram_loss_1: 20.66242790222168
+INFO:__main__:vgg_gram_loss_2: 21.898353576660156
+INFO:__main__:vgg_gram_loss_3: 21.246999740600586
+INFO:__main__:vgg_gram_loss_4: 84.6812973022461
+INFO:__main__:vgg_gram_loss_5: 0.1856197565793991
+INFO:__main__:vgg_loss_0: 11.673689842224121
+INFO:__main__:vgg_loss_1: 30.411474227905273
+INFO:__main__:vgg_loss_2: 41.39936828613281
+INFO:__main__:vgg_loss_3: 29.173614501953125
+INFO:__main__:vgg_loss_4: 50.74263381958008
+INFO:__main__:vgg_loss_5: 1.6937453746795654
+INFO:__main__:validation_loss: 1977.4515380859375
+INFO:__main__:global_step: 69501
+INFO:__main__:kl_loss: 134.11952209472656
+INFO:__main__:kl_weight: 0.7800002098083496
+INFO:__main__:likelihood_loss: 1518.17626953125
+INFO:__main__:loss: 1622.78955078125
+INFO:__main__:vgg_gram_loss_0: 6.403439521789551
+INFO:__main__:vgg_gram_loss_1: 23.630760192871094
+INFO:__main__:vgg_gram_loss_2: 25.288049697875977
+INFO:__main__:vgg_gram_loss_3: 18.252885818481445
+INFO:__main__:vgg_gram_loss_4: 69.32908630371094
+INFO:__main__:vgg_gram_loss_5: 0.14441928267478943
+INFO:__main__:vgg_loss_0: 12.037620544433594
+INFO:__main__:vgg_loss_1: 30.384794235229492
+INFO:__main__:vgg_loss_2: 40.51081466674805
+INFO:__main__:vgg_loss_3: 27.982465744018555
+INFO:__main__:vgg_loss_4: 48.102970123291016
+INFO:__main__:vgg_loss_5: 1.5679389238357544
+INFO:__main__:validation_loss: 1976.461181640625
+INFO:__main__:global_step: 69751
+INFO:__main__:kl_loss: 101.4223403930664
+INFO:__main__:kl_weight: 0.7900002002716064
+INFO:__main__:likelihood_loss: 1594.7515869140625
+INFO:__main__:loss: 1674.875244140625
+INFO:__main__:vgg_gram_loss_0: 6.184321880340576
+INFO:__main__:vgg_gram_loss_1: 19.901065826416016
+INFO:__main__:vgg_gram_loss_2: 22.50301170349121
+INFO:__main__:vgg_gram_loss_3: 17.64005470275879
+INFO:__main__:vgg_gram_loss_4: 78.64226531982422
+INFO:__main__:vgg_gram_loss_5: 0.19677816331386566
+INFO:__main__:vgg_loss_0: 12.225306510925293
+INFO:__main__:vgg_loss_1: 30.046615600585938
+INFO:__main__:vgg_loss_2: 44.126468658447266
+INFO:__main__:vgg_loss_3: 31.57122230529785
+INFO:__main__:vgg_loss_4: 54.11068344116211
+INFO:__main__:vgg_loss_5: 1.8025399446487427
+INFO:__main__:validation_loss: 1647.0897216796875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 70001
+INFO:__main__:kl_loss: 114.62054443359375
+INFO:__main__:kl_weight: 0.8000001907348633
+INFO:__main__:likelihood_loss: 1259.611083984375
+INFO:__main__:loss: 1351.3074951171875
+INFO:__main__:vgg_gram_loss_0: 11.280505180358887
+INFO:__main__:vgg_gram_loss_1: 17.14234161376953
+INFO:__main__:vgg_gram_loss_2: 17.81877899169922
+INFO:__main__:vgg_gram_loss_3: 13.883241653442383
+INFO:__main__:vgg_gram_loss_4: 60.410888671875
+INFO:__main__:vgg_gram_loss_5: 0.1228264570236206
+INFO:__main__:vgg_loss_0: 10.088823318481445
+INFO:__main__:vgg_loss_1: 22.671152114868164
+INFO:__main__:vgg_loss_2: 32.982757568359375
+INFO:__main__:vgg_loss_3: 23.601299285888672
+INFO:__main__:vgg_loss_4: 40.587345123291016
+INFO:__main__:vgg_loss_5: 1.33225679397583
+INFO:__main__:validation_loss: 1970.864990234375
+INFO:__main__:global_step: 70251
+INFO:__main__:kl_loss: 122.21644592285156
+INFO:__main__:kl_weight: 0.8100001811981201
+INFO:__main__:likelihood_loss: 1489.11279296875
+INFO:__main__:loss: 1588.108154296875
+INFO:__main__:vgg_gram_loss_0: 5.7129950523376465
+INFO:__main__:vgg_gram_loss_1: 19.401472091674805
+INFO:__main__:vgg_gram_loss_2: 22.117023468017578
+INFO:__main__:vgg_gram_loss_3: 17.118972778320312
+INFO:__main__:vgg_gram_loss_4: 70.82239532470703
+INFO:__main__:vgg_gram_loss_5: 0.1722775101661682
+INFO:__main__:vgg_loss_0: 11.55367660522461
+INFO:__main__:vgg_loss_1: 28.541751861572266
+INFO:__main__:vgg_loss_2: 40.3692626953125
+INFO:__main__:vgg_loss_3: 29.07376480102539
+INFO:__main__:vgg_loss_4: 51.23691940307617
+INFO:__main__:vgg_loss_5: 1.7020846605300903
+INFO:__main__:validation_loss: 1889.73486328125
+INFO:__main__:global_step: 70501
+INFO:__main__:kl_loss: 100.42434692382812
+INFO:__main__:kl_weight: 0.820000171661377
+INFO:__main__:likelihood_loss: 1476.46875
+INFO:__main__:loss: 1558.8167724609375
+INFO:__main__:vgg_gram_loss_0: 7.895826816558838
+INFO:__main__:vgg_gram_loss_1: 18.152254104614258
+INFO:__main__:vgg_gram_loss_2: 19.197431564331055
+INFO:__main__:vgg_gram_loss_3: 16.37513542175293
+INFO:__main__:vgg_gram_loss_4: 74.21931457519531
+INFO:__main__:vgg_gram_loss_5: 0.17780187726020813
+INFO:__main__:vgg_loss_0: 11.138297080993652
+INFO:__main__:vgg_loss_1: 28.024188995361328
+INFO:__main__:vgg_loss_2: 39.97074508666992
+INFO:__main__:vgg_loss_3: 28.34901237487793
+INFO:__main__:vgg_loss_4: 50.11932373046875
+INFO:__main__:vgg_loss_5: 1.6744214296340942
+INFO:__main__:validation_loss: 1503.5784912109375
+INFO:__main__:global_step: 70751
+INFO:__main__:kl_loss: 110.1641845703125
+INFO:__main__:kl_weight: 0.8300001621246338
+INFO:__main__:likelihood_loss: 1849.4932861328125
+INFO:__main__:loss: 1940.9295654296875
+INFO:__main__:vgg_gram_loss_0: 5.697214603424072
+INFO:__main__:vgg_gram_loss_1: 29.907506942749023
+INFO:__main__:vgg_gram_loss_2: 25.6283016204834
+INFO:__main__:vgg_gram_loss_3: 26.94020652770996
+INFO:__main__:vgg_gram_loss_4: 94.22338104248047
+INFO:__main__:vgg_gram_loss_5: 0.1919695883989334
+INFO:__main__:vgg_loss_0: 14.222413063049316
+INFO:__main__:vgg_loss_1: 35.892452239990234
+INFO:__main__:vgg_loss_2: 46.05123519897461
+INFO:__main__:vgg_loss_3: 32.38946533203125
+INFO:__main__:vgg_loss_4: 56.8395881652832
+INFO:__main__:vgg_loss_5: 1.9149153232574463
+INFO:__main__:validation_loss: 2107.130615234375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 71001
+INFO:__main__:kl_loss: 123.37052917480469
+INFO:__main__:kl_weight: 0.8400001525878906
+INFO:__main__:likelihood_loss: 1531.439453125
+INFO:__main__:loss: 1635.0706787109375
+INFO:__main__:vgg_gram_loss_0: 7.619284152984619
+INFO:__main__:vgg_gram_loss_1: 18.909658432006836
+INFO:__main__:vgg_gram_loss_2: 19.578994750976562
+INFO:__main__:vgg_gram_loss_3: 16.31967544555664
+INFO:__main__:vgg_gram_loss_4: 78.66064453125
+INFO:__main__:vgg_gram_loss_5: 0.17580434679985046
+INFO:__main__:vgg_loss_0: 11.272994995117188
+INFO:__main__:vgg_loss_1: 27.591440200805664
+INFO:__main__:vgg_loss_2: 40.4105224609375
+INFO:__main__:vgg_loss_3: 30.06654167175293
+INFO:__main__:vgg_loss_4: 53.899295806884766
+INFO:__main__:vgg_loss_5: 1.7830345630645752
+INFO:__main__:validation_loss: 2099.07861328125
+INFO:__main__:global_step: 71251
+INFO:__main__:kl_loss: 114.05953979492188
+INFO:__main__:kl_weight: 0.8500001430511475
+INFO:__main__:likelihood_loss: 1501.0948486328125
+INFO:__main__:loss: 1598.04541015625
+INFO:__main__:vgg_gram_loss_0: 7.028250217437744
+INFO:__main__:vgg_gram_loss_1: 18.595855712890625
+INFO:__main__:vgg_gram_loss_2: 19.473913192749023
+INFO:__main__:vgg_gram_loss_3: 16.306814193725586
+INFO:__main__:vgg_gram_loss_4: 78.75574493408203
+INFO:__main__:vgg_gram_loss_5: 0.18558269739151
+INFO:__main__:vgg_loss_0: 10.752283096313477
+INFO:__main__:vgg_loss_1: 25.481149673461914
+INFO:__main__:vgg_loss_2: 39.071964263916016
+INFO:__main__:vgg_loss_3: 29.390300750732422
+INFO:__main__:vgg_loss_4: 53.36310577392578
+INFO:__main__:vgg_loss_5: 1.8139877319335938
+INFO:__main__:validation_loss: 2232.254638671875
+INFO:__main__:global_step: 71501
+INFO:__main__:kl_loss: 100.27262878417969
+INFO:__main__:kl_weight: 0.8600001335144043
+INFO:__main__:likelihood_loss: 1220.089111328125
+INFO:__main__:loss: 1306.3236083984375
+INFO:__main__:vgg_gram_loss_0: 4.074394226074219
+INFO:__main__:vgg_gram_loss_1: 12.995302200317383
+INFO:__main__:vgg_gram_loss_2: 16.70504379272461
+INFO:__main__:vgg_gram_loss_3: 13.237479209899902
+INFO:__main__:vgg_gram_loss_4: 62.839412689208984
+INFO:__main__:vgg_gram_loss_5: 0.13876856863498688
+INFO:__main__:vgg_loss_0: 9.493607521057129
+INFO:__main__:vgg_loss_1: 23.230228424072266
+INFO:__main__:vgg_loss_2: 34.059608459472656
+INFO:__main__:vgg_loss_3: 24.03904151916504
+INFO:__main__:vgg_loss_4: 41.82756423950195
+INFO:__main__:vgg_loss_5: 1.3773560523986816
+INFO:__main__:validation_loss: 1982.9581298828125
+INFO:__main__:global_step: 71751
+INFO:__main__:kl_loss: 113.09410858154297
+INFO:__main__:kl_weight: 0.8700001239776611
+INFO:__main__:likelihood_loss: 1513.96337890625
+INFO:__main__:loss: 1612.355224609375
+INFO:__main__:vgg_gram_loss_0: 7.728318214416504
+INFO:__main__:vgg_gram_loss_1: 16.285978317260742
+INFO:__main__:vgg_gram_loss_2: 18.800724029541016
+INFO:__main__:vgg_gram_loss_3: 16.16196060180664
+INFO:__main__:vgg_gram_loss_4: 78.34253692626953
+INFO:__main__:vgg_gram_loss_5: 0.19268853962421417
+INFO:__main__:vgg_loss_0: 11.200359344482422
+INFO:__main__:vgg_loss_1: 27.516067504882812
+INFO:__main__:vgg_loss_2: 41.83439636230469
+INFO:__main__:vgg_loss_3: 30.327730178833008
+INFO:__main__:vgg_loss_4: 52.63300323486328
+INFO:__main__:vgg_loss_5: 1.768884301185608
+INFO:__main__:validation_loss: 1924.7484130859375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 72001
+INFO:__main__:kl_loss: 112.17937469482422
+INFO:__main__:kl_weight: 0.880000114440918
+INFO:__main__:likelihood_loss: 1309.4434814453125
+INFO:__main__:loss: 1408.161376953125
+INFO:__main__:vgg_gram_loss_0: 6.0181403160095215
+INFO:__main__:vgg_gram_loss_1: 15.1677885055542
+INFO:__main__:vgg_gram_loss_2: 17.82264518737793
+INFO:__main__:vgg_gram_loss_3: 13.64476490020752
+INFO:__main__:vgg_gram_loss_4: 66.82454681396484
+INFO:__main__:vgg_gram_loss_5: 0.16753584146499634
+INFO:__main__:vgg_loss_0: 10.165095329284668
+INFO:__main__:vgg_loss_1: 24.31793975830078
+INFO:__main__:vgg_loss_2: 35.764686584472656
+INFO:__main__:vgg_loss_3: 25.912620544433594
+INFO:__main__:vgg_loss_4: 44.63575744628906
+INFO:__main__:vgg_loss_5: 1.447156310081482
+INFO:__main__:validation_loss: 1988.7105712890625
+INFO:__main__:global_step: 72251
+INFO:__main__:kl_loss: 97.12775421142578
+INFO:__main__:kl_weight: 0.8900001049041748
+INFO:__main__:likelihood_loss: 1309.65673828125
+INFO:__main__:loss: 1396.1004638671875
+INFO:__main__:vgg_gram_loss_0: 6.9812140464782715
+INFO:__main__:vgg_gram_loss_1: 14.923712730407715
+INFO:__main__:vgg_gram_loss_2: 17.462730407714844
+INFO:__main__:vgg_gram_loss_3: 14.518150329589844
+INFO:__main__:vgg_gram_loss_4: 63.82509994506836
+INFO:__main__:vgg_gram_loss_5: 0.14979802072048187
+INFO:__main__:vgg_loss_0: 10.59226131439209
+INFO:__main__:vgg_loss_1: 24.807628631591797
+INFO:__main__:vgg_loss_2: 36.32770538330078
+INFO:__main__:vgg_loss_3: 25.696069717407227
+INFO:__main__:vgg_loss_4: 45.144996643066406
+INFO:__main__:vgg_loss_5: 1.5019564628601074
+INFO:__main__:validation_loss: 1940.3394775390625
+INFO:__main__:global_step: 72501
+INFO:__main__:kl_loss: 99.95515441894531
+INFO:__main__:kl_weight: 0.9000000953674316
+INFO:__main__:likelihood_loss: 1565.877685546875
+INFO:__main__:loss: 1655.8372802734375
+INFO:__main__:vgg_gram_loss_0: 5.93556022644043
+INFO:__main__:vgg_gram_loss_1: 16.67200469970703
+INFO:__main__:vgg_gram_loss_2: 23.253843307495117
+INFO:__main__:vgg_gram_loss_3: 18.802677154541016
+INFO:__main__:vgg_gram_loss_4: 80.59404754638672
+INFO:__main__:vgg_gram_loss_5: 0.17576146125793457
+INFO:__main__:vgg_loss_0: 11.554752349853516
+INFO:__main__:vgg_loss_1: 29.07073211669922
+INFO:__main__:vgg_loss_2: 42.50658416748047
+INFO:__main__:vgg_loss_3: 30.3881778717041
+INFO:__main__:vgg_loss_4: 52.49169158935547
+INFO:__main__:vgg_loss_5: 1.7296836376190186
+INFO:__main__:validation_loss: 2278.136474609375
+INFO:__main__:global_step: 72751
+INFO:__main__:kl_loss: 113.69381713867188
+INFO:__main__:kl_weight: 0.9100000858306885
+INFO:__main__:likelihood_loss: 1682.2967529296875
+INFO:__main__:loss: 1785.7581787109375
+INFO:__main__:vgg_gram_loss_0: 9.08698558807373
+INFO:__main__:vgg_gram_loss_1: 22.309814453125
+INFO:__main__:vgg_gram_loss_2: 26.631031036376953
+INFO:__main__:vgg_gram_loss_3: 22.305757522583008
+INFO:__main__:vgg_gram_loss_4: 84.04031372070312
+INFO:__main__:vgg_gram_loss_5: 0.17333360016345978
+INFO:__main__:vgg_loss_0: 12.1488676071167
+INFO:__main__:vgg_loss_1: 30.9793758392334
+INFO:__main__:vgg_loss_2: 43.765037536621094
+INFO:__main__:vgg_loss_3: 30.647972106933594
+INFO:__main__:vgg_loss_4: 52.655250549316406
+INFO:__main__:vgg_loss_5: 1.7155909538269043
+INFO:__main__:validation_loss: 2161.9501953125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 73001
+INFO:__main__:kl_loss: 119.75785064697266
+INFO:__main__:kl_weight: 0.9200000762939453
+INFO:__main__:likelihood_loss: 1499.76416015625
+INFO:__main__:loss: 1609.94140625
+INFO:__main__:vgg_gram_loss_0: 5.143685340881348
+INFO:__main__:vgg_gram_loss_1: 18.755769729614258
+INFO:__main__:vgg_gram_loss_2: 24.124832153320312
+INFO:__main__:vgg_gram_loss_3: 18.614320755004883
+INFO:__main__:vgg_gram_loss_4: 74.55846405029297
+INFO:__main__:vgg_gram_loss_5: 0.16492317616939545
+INFO:__main__:vgg_loss_0: 10.954079627990723
+INFO:__main__:vgg_loss_1: 28.294960021972656
+INFO:__main__:vgg_loss_2: 40.197471618652344
+INFO:__main__:vgg_loss_3: 29.055871963500977
+INFO:__main__:vgg_loss_4: 48.49783706665039
+INFO:__main__:vgg_loss_5: 1.5905847549438477
+INFO:__main__:validation_loss: 2446.968994140625
+INFO:__main__:global_step: 73251
+INFO:__main__:kl_loss: 101.9072494506836
+INFO:__main__:kl_weight: 0.9300000667572021
+INFO:__main__:likelihood_loss: 1496.933837890625
+INFO:__main__:loss: 1591.7076416015625
+INFO:__main__:vgg_gram_loss_0: 7.158200263977051
+INFO:__main__:vgg_gram_loss_1: 19.177204132080078
+INFO:__main__:vgg_gram_loss_2: 20.175968170166016
+INFO:__main__:vgg_gram_loss_3: 17.971181869506836
+INFO:__main__:vgg_gram_loss_4: 71.14128112792969
+INFO:__main__:vgg_gram_loss_5: 0.17106644809246063
+INFO:__main__:vgg_loss_0: 12.636214256286621
+INFO:__main__:vgg_loss_1: 31.10297393798828
+INFO:__main__:vgg_loss_2: 41.990264892578125
+INFO:__main__:vgg_loss_3: 28.953113555908203
+INFO:__main__:vgg_loss_4: 47.32289123535156
+INFO:__main__:vgg_loss_5: 1.5863882303237915
+INFO:__main__:validation_loss: 2045.115478515625
+INFO:__main__:global_step: 73501
+INFO:__main__:kl_loss: 109.71762084960938
+INFO:__main__:kl_weight: 0.940000057220459
+INFO:__main__:likelihood_loss: 1646.0426025390625
+INFO:__main__:loss: 1749.1771240234375
+INFO:__main__:vgg_gram_loss_0: 8.803943634033203
+INFO:__main__:vgg_gram_loss_1: 21.761241912841797
+INFO:__main__:vgg_gram_loss_2: 26.873910903930664
+INFO:__main__:vgg_gram_loss_3: 19.72337532043457
+INFO:__main__:vgg_gram_loss_4: 82.79439544677734
+INFO:__main__:vgg_gram_loss_5: 0.17672832310199738
+INFO:__main__:vgg_loss_0: 11.911394119262695
+INFO:__main__:vgg_loss_1: 28.469045639038086
+INFO:__main__:vgg_loss_2: 41.5036506652832
+INFO:__main__:vgg_loss_3: 30.807981491088867
+INFO:__main__:vgg_loss_4: 54.57335662841797
+INFO:__main__:vgg_loss_5: 1.8094980716705322
+INFO:__main__:validation_loss: 2040.6015625
+INFO:__main__:global_step: 73751
+INFO:__main__:kl_loss: 113.99014282226562
+INFO:__main__:kl_weight: 0.9500000476837158
+INFO:__main__:likelihood_loss: 1590.9906005859375
+INFO:__main__:loss: 1699.28125
+INFO:__main__:vgg_gram_loss_0: 5.220655918121338
+INFO:__main__:vgg_gram_loss_1: 25.686182022094727
+INFO:__main__:vgg_gram_loss_2: 26.844196319580078
+INFO:__main__:vgg_gram_loss_3: 19.788684844970703
+INFO:__main__:vgg_gram_loss_4: 80.70267486572266
+INFO:__main__:vgg_gram_loss_5: 0.18536019325256348
+INFO:__main__:vgg_loss_0: 10.139582633972168
+INFO:__main__:vgg_loss_1: 26.16705322265625
+INFO:__main__:vgg_loss_2: 40.3869743347168
+INFO:__main__:vgg_loss_3: 29.833858489990234
+INFO:__main__:vgg_loss_4: 51.54114532470703
+INFO:__main__:vgg_loss_5: 1.7017500400543213
+INFO:__main__:validation_loss: 1680.5662841796875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 74001
+INFO:__main__:kl_loss: 125.37686920166016
+INFO:__main__:kl_weight: 0.9600000381469727
+INFO:__main__:likelihood_loss: 1664.7703857421875
+INFO:__main__:loss: 1785.1322021484375
+INFO:__main__:vgg_gram_loss_0: 5.143579959869385
+INFO:__main__:vgg_gram_loss_1: 20.92984390258789
+INFO:__main__:vgg_gram_loss_2: 26.126178741455078
+INFO:__main__:vgg_gram_loss_3: 19.386747360229492
+INFO:__main__:vgg_gram_loss_4: 83.82691955566406
+INFO:__main__:vgg_gram_loss_5: 0.1983432024717331
+INFO:__main__:vgg_loss_0: 12.82373332977295
+INFO:__main__:vgg_loss_1: 33.68911361694336
+INFO:__main__:vgg_loss_2: 45.28364944458008
+INFO:__main__:vgg_loss_3: 31.59891700744629
+INFO:__main__:vgg_loss_4: 52.19144821166992
+INFO:__main__:vgg_loss_5: 1.7555961608886719
+INFO:__main__:validation_loss: 2110.444091796875
+INFO:__main__:global_step: 74251
+INFO:__main__:kl_loss: 102.18760681152344
+INFO:__main__:kl_weight: 0.9700000286102295
+INFO:__main__:likelihood_loss: 1639.598388671875
+INFO:__main__:loss: 1738.7203369140625
+INFO:__main__:vgg_gram_loss_0: 9.353968620300293
+INFO:__main__:vgg_gram_loss_1: 20.314062118530273
+INFO:__main__:vgg_gram_loss_2: 22.032384872436523
+INFO:__main__:vgg_gram_loss_3: 17.38332176208496
+INFO:__main__:vgg_gram_loss_4: 79.24800872802734
+INFO:__main__:vgg_gram_loss_5: 0.17939458787441254
+INFO:__main__:vgg_loss_0: 13.099538803100586
+INFO:__main__:vgg_loss_1: 31.539186477661133
+INFO:__main__:vgg_loss_2: 44.58019256591797
+INFO:__main__:vgg_loss_3: 32.09550476074219
+INFO:__main__:vgg_loss_4: 56.25748062133789
+INFO:__main__:vgg_loss_5: 1.8366080522537231
+INFO:__main__:validation_loss: 1797.7974853515625
+INFO:__main__:global_step: 74501
+INFO:__main__:kl_loss: 101.51658630371094
+INFO:__main__:kl_weight: 0.9800000190734863
+INFO:__main__:likelihood_loss: 1649.2950439453125
+INFO:__main__:loss: 1748.78125
+INFO:__main__:vgg_gram_loss_0: 7.402290344238281
+INFO:__main__:vgg_gram_loss_1: 19.013675689697266
+INFO:__main__:vgg_gram_loss_2: 21.814207077026367
+INFO:__main__:vgg_gram_loss_3: 28.09962272644043
+INFO:__main__:vgg_gram_loss_4: 89.76759338378906
+INFO:__main__:vgg_gram_loss_5: 0.17110000550746918
+INFO:__main__:vgg_loss_0: 12.822097778320312
+INFO:__main__:vgg_loss_1: 29.958127975463867
+INFO:__main__:vgg_loss_2: 40.26081085205078
+INFO:__main__:vgg_loss_3: 29.256885528564453
+INFO:__main__:vgg_loss_4: 49.66324996948242
+INFO:__main__:vgg_loss_5: 1.6293420791625977
+INFO:__main__:validation_loss: 1908.570556640625
+INFO:__main__:global_step: 74751
+INFO:__main__:kl_loss: 104.74473571777344
+INFO:__main__:kl_weight: 0.9900000095367432
+INFO:__main__:likelihood_loss: 1764.0521240234375
+INFO:__main__:loss: 1867.7493896484375
+INFO:__main__:vgg_gram_loss_0: 5.571436405181885
+INFO:__main__:vgg_gram_loss_1: 29.543439865112305
+INFO:__main__:vgg_gram_loss_2: 27.647619247436523
+INFO:__main__:vgg_gram_loss_3: 23.159008026123047
+INFO:__main__:vgg_gram_loss_4: 87.82096862792969
+INFO:__main__:vgg_gram_loss_5: 0.1866638958454132
+INFO:__main__:vgg_loss_0: 12.763458251953125
+INFO:__main__:vgg_loss_1: 32.14930725097656
+INFO:__main__:vgg_loss_2: 42.320499420166016
+INFO:__main__:vgg_loss_3: 31.915225982666016
+INFO:__main__:vgg_loss_4: 57.86028289794922
+INFO:__main__:vgg_loss_5: 1.872523546218872
+INFO:__main__:validation_loss: 1926.8133544921875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 75001
+INFO:__main__:kl_loss: 100.13243103027344
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1677.982177734375
+INFO:__main__:loss: 1778.1146240234375
+INFO:__main__:vgg_gram_loss_0: 10.163580894470215
+INFO:__main__:vgg_gram_loss_1: 20.064346313476562
+INFO:__main__:vgg_gram_loss_2: 22.39666748046875
+INFO:__main__:vgg_gram_loss_3: 19.094619750976562
+INFO:__main__:vgg_gram_loss_4: 84.0228042602539
+INFO:__main__:vgg_gram_loss_5: 0.1875264048576355
+INFO:__main__:vgg_loss_0: 13.289976119995117
+INFO:__main__:vgg_loss_1: 32.814308166503906
+INFO:__main__:vgg_loss_2: 44.40785598754883
+INFO:__main__:vgg_loss_3: 31.80345344543457
+INFO:__main__:vgg_loss_4: 55.50514602661133
+INFO:__main__:vgg_loss_5: 1.846155047416687
+INFO:__main__:validation_loss: 1914.8616943359375
+INFO:__main__:global_step: 75251
+INFO:__main__:kl_loss: 93.90699768066406
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1595.639892578125
+INFO:__main__:loss: 1689.546875
+INFO:__main__:vgg_gram_loss_0: 5.344332218170166
+INFO:__main__:vgg_gram_loss_1: 20.589935302734375
+INFO:__main__:vgg_gram_loss_2: 25.127233505249023
+INFO:__main__:vgg_gram_loss_3: 19.319421768188477
+INFO:__main__:vgg_gram_loss_4: 77.65477752685547
+INFO:__main__:vgg_gram_loss_5: 0.1963854730129242
+INFO:__main__:vgg_loss_0: 12.007538795471191
+INFO:__main__:vgg_loss_1: 29.647506713867188
+INFO:__main__:vgg_loss_2: 42.942832946777344
+INFO:__main__:vgg_loss_3: 30.897146224975586
+INFO:__main__:vgg_loss_4: 53.51076126098633
+INFO:__main__:vgg_loss_5: 1.8901301622390747
+INFO:__main__:validation_loss: 1979.13427734375
+INFO:__main__:global_step: 75501
+INFO:__main__:kl_loss: 97.80059814453125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1691.90771484375
+INFO:__main__:loss: 1789.708251953125
+INFO:__main__:vgg_gram_loss_0: 10.093396186828613
+INFO:__main__:vgg_gram_loss_1: 22.157413482666016
+INFO:__main__:vgg_gram_loss_2: 23.613557815551758
+INFO:__main__:vgg_gram_loss_3: 20.20595359802246
+INFO:__main__:vgg_gram_loss_4: 86.68827056884766
+INFO:__main__:vgg_gram_loss_5: 0.18877571821212769
+INFO:__main__:vgg_loss_0: 12.619112014770508
+INFO:__main__:vgg_loss_1: 29.3277530670166
+INFO:__main__:vgg_loss_2: 42.06028747558594
+INFO:__main__:vgg_loss_3: 32.2247200012207
+INFO:__main__:vgg_loss_4: 57.30364227294922
+INFO:__main__:vgg_loss_5: 1.8986399173736572
+INFO:__main__:validation_loss: 1890.991943359375
+INFO:__main__:global_step: 75751
+INFO:__main__:kl_loss: 109.33937072753906
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1598.21630859375
+INFO:__main__:loss: 1707.5556640625
+INFO:__main__:vgg_gram_loss_0: 7.216796875
+INFO:__main__:vgg_gram_loss_1: 18.278650283813477
+INFO:__main__:vgg_gram_loss_2: 21.29659080505371
+INFO:__main__:vgg_gram_loss_3: 17.454660415649414
+INFO:__main__:vgg_gram_loss_4: 76.95777893066406
+INFO:__main__:vgg_gram_loss_5: 0.1859695166349411
+INFO:__main__:vgg_loss_0: 13.163182258605957
+INFO:__main__:vgg_loss_1: 30.361055374145508
+INFO:__main__:vgg_loss_2: 44.17300796508789
+INFO:__main__:vgg_loss_3: 32.27228927612305
+INFO:__main__:vgg_loss_4: 56.428749084472656
+INFO:__main__:vgg_loss_5: 1.8545019626617432
+INFO:__main__:validation_loss: 1732.8148193359375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 76001
+INFO:__main__:kl_loss: 103.81758117675781
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1586.361328125
+INFO:__main__:loss: 1690.178955078125
+INFO:__main__:vgg_gram_loss_0: 7.6762213706970215
+INFO:__main__:vgg_gram_loss_1: 18.554441452026367
+INFO:__main__:vgg_gram_loss_2: 24.411039352416992
+INFO:__main__:vgg_gram_loss_3: 19.433027267456055
+INFO:__main__:vgg_gram_loss_4: 81.15123748779297
+INFO:__main__:vgg_gram_loss_5: 0.17046280205249786
+INFO:__main__:vgg_loss_0: 11.782315254211426
+INFO:__main__:vgg_loss_1: 27.911771774291992
+INFO:__main__:vgg_loss_2: 42.02341079711914
+INFO:__main__:vgg_loss_3: 30.431303024291992
+INFO:__main__:vgg_loss_4: 52.03547668457031
+INFO:__main__:vgg_loss_5: 1.691562294960022
+INFO:__main__:validation_loss: 2026.08984375
+INFO:__main__:global_step: 76251
+INFO:__main__:kl_loss: 104.38290405273438
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1316.176513671875
+INFO:__main__:loss: 1420.5594482421875
+INFO:__main__:vgg_gram_loss_0: 4.116774082183838
+INFO:__main__:vgg_gram_loss_1: 12.36014461517334
+INFO:__main__:vgg_gram_loss_2: 18.176054000854492
+INFO:__main__:vgg_gram_loss_3: 14.217645645141602
+INFO:__main__:vgg_gram_loss_4: 67.14892578125
+INFO:__main__:vgg_gram_loss_5: 0.16229002177715302
+INFO:__main__:vgg_loss_0: 10.36562442779541
+INFO:__main__:vgg_loss_1: 24.966463088989258
+INFO:__main__:vgg_loss_2: 37.10487365722656
+INFO:__main__:vgg_loss_3: 26.90335464477539
+INFO:__main__:vgg_loss_4: 46.18036651611328
+INFO:__main__:vgg_loss_5: 1.5327625274658203
+INFO:__main__:validation_loss: 1801.510498046875
+INFO:__main__:global_step: 76501
+INFO:__main__:kl_loss: 113.70652770996094
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1440.7744140625
+INFO:__main__:loss: 1554.48095703125
+INFO:__main__:vgg_gram_loss_0: 4.452088832855225
+INFO:__main__:vgg_gram_loss_1: 15.967294692993164
+INFO:__main__:vgg_gram_loss_2: 18.8958740234375
+INFO:__main__:vgg_gram_loss_3: 15.105307579040527
+INFO:__main__:vgg_gram_loss_4: 72.4871597290039
+INFO:__main__:vgg_gram_loss_5: 0.17030729353427887
+INFO:__main__:vgg_loss_0: 10.69140911102295
+INFO:__main__:vgg_loss_1: 26.644725799560547
+INFO:__main__:vgg_loss_2: 39.99558639526367
+INFO:__main__:vgg_loss_3: 29.811471939086914
+INFO:__main__:vgg_loss_4: 52.17962646484375
+INFO:__main__:vgg_loss_5: 1.7540267705917358
+INFO:__main__:validation_loss: 1987.474365234375
+INFO:__main__:global_step: 76751
+INFO:__main__:kl_loss: 95.03602600097656
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1320.235595703125
+INFO:__main__:loss: 1415.2716064453125
+INFO:__main__:vgg_gram_loss_0: 8.905549049377441
+INFO:__main__:vgg_gram_loss_1: 17.867881774902344
+INFO:__main__:vgg_gram_loss_2: 18.062788009643555
+INFO:__main__:vgg_gram_loss_3: 15.119465827941895
+INFO:__main__:vgg_gram_loss_4: 65.11434173583984
+INFO:__main__:vgg_gram_loss_5: 0.14481575787067413
+INFO:__main__:vgg_loss_0: 9.570205688476562
+INFO:__main__:vgg_loss_1: 24.211755752563477
+INFO:__main__:vgg_loss_2: 35.236228942871094
+INFO:__main__:vgg_loss_3: 24.88789176940918
+INFO:__main__:vgg_loss_4: 43.45017623901367
+INFO:__main__:vgg_loss_5: 1.4760310649871826
+INFO:__main__:validation_loss: 2242.143798828125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 77001
+INFO:__main__:kl_loss: 99.5011978149414
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1474.970458984375
+INFO:__main__:loss: 1574.4716796875
+INFO:__main__:vgg_gram_loss_0: 5.013479709625244
+INFO:__main__:vgg_gram_loss_1: 13.940515518188477
+INFO:__main__:vgg_gram_loss_2: 22.664884567260742
+INFO:__main__:vgg_gram_loss_3: 16.834014892578125
+INFO:__main__:vgg_gram_loss_4: 72.3034896850586
+INFO:__main__:vgg_gram_loss_5: 0.16409777104854584
+INFO:__main__:vgg_loss_0: 11.404167175292969
+INFO:__main__:vgg_loss_1: 27.72151756286621
+INFO:__main__:vgg_loss_2: 40.281005859375
+INFO:__main__:vgg_loss_3: 29.359331130981445
+INFO:__main__:vgg_loss_4: 53.53948211669922
+INFO:__main__:vgg_loss_5: 1.7681152820587158
+INFO:__main__:validation_loss: 1739.6978759765625
+INFO:__main__:global_step: 77251
+INFO:__main__:kl_loss: 93.35214233398438
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1837.0057373046875
+INFO:__main__:loss: 1930.35791015625
+INFO:__main__:vgg_gram_loss_0: 3.5470542907714844
+INFO:__main__:vgg_gram_loss_1: 19.482412338256836
+INFO:__main__:vgg_gram_loss_2: 24.20562171936035
+INFO:__main__:vgg_gram_loss_3: 18.212350845336914
+INFO:__main__:vgg_gram_loss_4: 81.92411041259766
+INFO:__main__:vgg_gram_loss_5: 0.19002985954284668
+INFO:__main__:vgg_loss_0: 17.35188865661621
+INFO:__main__:vgg_loss_1: 42.162376403808594
+INFO:__main__:vgg_loss_2: 55.32271957397461
+INFO:__main__:vgg_loss_3: 37.95719528198242
+INFO:__main__:vgg_loss_4: 64.96937561035156
+INFO:__main__:vgg_loss_5: 2.076005458831787
+INFO:__main__:validation_loss: 2063.02197265625
+INFO:__main__:global_step: 77501
+INFO:__main__:kl_loss: 98.62416076660156
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1351.188720703125
+INFO:__main__:loss: 1449.8128662109375
+INFO:__main__:vgg_gram_loss_0: 3.9217097759246826
+INFO:__main__:vgg_gram_loss_1: 18.386978149414062
+INFO:__main__:vgg_gram_loss_2: 17.819812774658203
+INFO:__main__:vgg_gram_loss_3: 13.708714485168457
+INFO:__main__:vgg_gram_loss_4: 62.57135009765625
+INFO:__main__:vgg_gram_loss_5: 0.14589530229568481
+INFO:__main__:vgg_loss_0: 11.606657981872559
+INFO:__main__:vgg_loss_1: 28.303672790527344
+INFO:__main__:vgg_loss_2: 38.20246505737305
+INFO:__main__:vgg_loss_3: 27.203811645507812
+INFO:__main__:vgg_loss_4: 46.85149383544922
+INFO:__main__:vgg_loss_5: 1.515182375907898
+INFO:__main__:validation_loss: 2588.727783203125
+INFO:__main__:global_step: 77751
+INFO:__main__:kl_loss: 95.46920776367188
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1571.1240234375
+INFO:__main__:loss: 1666.59326171875
+INFO:__main__:vgg_gram_loss_0: 5.44347620010376
+INFO:__main__:vgg_gram_loss_1: 17.16338539123535
+INFO:__main__:vgg_gram_loss_2: 20.288297653198242
+INFO:__main__:vgg_gram_loss_3: 17.379610061645508
+INFO:__main__:vgg_gram_loss_4: 81.77371215820312
+INFO:__main__:vgg_gram_loss_5: 0.18628744781017303
+INFO:__main__:vgg_loss_0: 12.258504867553711
+INFO:__main__:vgg_loss_1: 30.573843002319336
+INFO:__main__:vgg_loss_2: 42.83974838256836
+INFO:__main__:vgg_loss_3: 30.539752960205078
+INFO:__main__:vgg_loss_4: 53.980838775634766
+INFO:__main__:vgg_loss_5: 1.797322154045105
+INFO:__main__:validation_loss: 1623.26416015625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 78001
+INFO:__main__:kl_loss: 99.60126495361328
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1897.071533203125
+INFO:__main__:loss: 1996.6728515625
+INFO:__main__:vgg_gram_loss_0: 6.299485206604004
+INFO:__main__:vgg_gram_loss_1: 22.75691032409668
+INFO:__main__:vgg_gram_loss_2: 31.383136749267578
+INFO:__main__:vgg_gram_loss_3: 23.634672164916992
+INFO:__main__:vgg_gram_loss_4: 97.5406265258789
+INFO:__main__:vgg_gram_loss_5: 0.20283500850200653
+INFO:__main__:vgg_loss_0: 14.569567680358887
+INFO:__main__:vgg_loss_1: 33.569129943847656
+INFO:__main__:vgg_loss_2: 50.29151916503906
+INFO:__main__:vgg_loss_3: 35.87866973876953
+INFO:__main__:vgg_loss_4: 61.29372024536133
+INFO:__main__:vgg_loss_5: 1.9939991235733032
+INFO:__main__:validation_loss: 2132.142333984375
+INFO:__main__:global_step: 78251
+INFO:__main__:kl_loss: 109.67769622802734
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1434.6954345703125
+INFO:__main__:loss: 1544.3731689453125
+INFO:__main__:vgg_gram_loss_0: 6.363452911376953
+INFO:__main__:vgg_gram_loss_1: 17.724069595336914
+INFO:__main__:vgg_gram_loss_2: 18.867610931396484
+INFO:__main__:vgg_gram_loss_3: 15.112489700317383
+INFO:__main__:vgg_gram_loss_4: 72.76872253417969
+INFO:__main__:vgg_gram_loss_5: 0.1608508825302124
+INFO:__main__:vgg_loss_0: 10.462724685668945
+INFO:__main__:vgg_loss_1: 25.66767692565918
+INFO:__main__:vgg_loss_2: 39.69873809814453
+INFO:__main__:vgg_loss_3: 28.601001739501953
+INFO:__main__:vgg_loss_4: 49.9183235168457
+INFO:__main__:vgg_loss_5: 1.5934256315231323
+INFO:__main__:validation_loss: 2078.099609375
+INFO:__main__:global_step: 78501
+INFO:__main__:kl_loss: 90.05607604980469
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1633.7392578125
+INFO:__main__:loss: 1723.7952880859375
+INFO:__main__:vgg_gram_loss_0: 6.281897068023682
+INFO:__main__:vgg_gram_loss_1: 15.997424125671387
+INFO:__main__:vgg_gram_loss_2: 22.45941162109375
+INFO:__main__:vgg_gram_loss_3: 22.636030197143555
+INFO:__main__:vgg_gram_loss_4: 87.31340789794922
+INFO:__main__:vgg_gram_loss_5: 0.19724315404891968
+INFO:__main__:vgg_loss_0: 11.639867782592773
+INFO:__main__:vgg_loss_1: 29.515239715576172
+INFO:__main__:vgg_loss_2: 43.279415130615234
+INFO:__main__:vgg_loss_3: 30.972524642944336
+INFO:__main__:vgg_loss_4: 54.58340072631836
+INFO:__main__:vgg_loss_5: 1.8719711303710938
+INFO:__main__:validation_loss: 2089.758544921875
+INFO:__main__:global_step: 78751
+INFO:__main__:kl_loss: 105.66798400878906
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1415.455322265625
+INFO:__main__:loss: 1521.123291015625
+INFO:__main__:vgg_gram_loss_0: 6.239748477935791
+INFO:__main__:vgg_gram_loss_1: 13.505637168884277
+INFO:__main__:vgg_gram_loss_2: 21.595430374145508
+INFO:__main__:vgg_gram_loss_3: 17.695768356323242
+INFO:__main__:vgg_gram_loss_4: 72.23534393310547
+INFO:__main__:vgg_gram_loss_5: 0.16348953545093536
+INFO:__main__:vgg_loss_0: 10.687366485595703
+INFO:__main__:vgg_loss_1: 25.60257911682129
+INFO:__main__:vgg_loss_2: 37.14620590209961
+INFO:__main__:vgg_loss_3: 27.485403060913086
+INFO:__main__:vgg_loss_4: 49.14472198486328
+INFO:__main__:vgg_loss_5: 1.5893570184707642
+INFO:__main__:validation_loss: 1848.4595947265625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 79001
+INFO:__main__:kl_loss: 88.70341491699219
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1411.13037109375
+INFO:__main__:loss: 1499.833740234375
+INFO:__main__:vgg_gram_loss_0: 7.180247783660889
+INFO:__main__:vgg_gram_loss_1: 15.113978385925293
+INFO:__main__:vgg_gram_loss_2: 18.890268325805664
+INFO:__main__:vgg_gram_loss_3: 15.731734275817871
+INFO:__main__:vgg_gram_loss_4: 71.05663299560547
+INFO:__main__:vgg_gram_loss_5: 0.1550329178571701
+INFO:__main__:vgg_loss_0: 10.854331970214844
+INFO:__main__:vgg_loss_1: 24.631467819213867
+INFO:__main__:vgg_loss_2: 37.0428352355957
+INFO:__main__:vgg_loss_3: 28.17113494873047
+INFO:__main__:vgg_loss_4: 51.70978546142578
+INFO:__main__:vgg_loss_5: 1.6886168718338013
+INFO:__main__:validation_loss: 1800.5113525390625
+INFO:__main__:global_step: 79251
+INFO:__main__:kl_loss: 97.91909790039062
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1300.59375
+INFO:__main__:loss: 1398.5128173828125
+INFO:__main__:vgg_gram_loss_0: 5.042819976806641
+INFO:__main__:vgg_gram_loss_1: 13.215660095214844
+INFO:__main__:vgg_gram_loss_2: 17.104307174682617
+INFO:__main__:vgg_gram_loss_3: 13.668588638305664
+INFO:__main__:vgg_gram_loss_4: 66.84371185302734
+INFO:__main__:vgg_gram_loss_5: 0.16382956504821777
+INFO:__main__:vgg_loss_0: 9.83410358428955
+INFO:__main__:vgg_loss_1: 23.349958419799805
+INFO:__main__:vgg_loss_2: 35.2518196105957
+INFO:__main__:vgg_loss_3: 26.384859085083008
+INFO:__main__:vgg_loss_4: 47.64868927001953
+INFO:__main__:vgg_loss_5: 1.610411524772644
+INFO:__main__:validation_loss: 2151.598388671875
+INFO:__main__:global_step: 79501
+INFO:__main__:kl_loss: 88.11454772949219
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1443.957275390625
+INFO:__main__:loss: 1532.07177734375
+INFO:__main__:vgg_gram_loss_0: 4.510222911834717
+INFO:__main__:vgg_gram_loss_1: 13.820660591125488
+INFO:__main__:vgg_gram_loss_2: 18.376134872436523
+INFO:__main__:vgg_gram_loss_3: 15.558046340942383
+INFO:__main__:vgg_gram_loss_4: 75.92063903808594
+INFO:__main__:vgg_gram_loss_5: 0.17562487721443176
+INFO:__main__:vgg_loss_0: 10.981514930725098
+INFO:__main__:vgg_loss_1: 27.408227920532227
+INFO:__main__:vgg_loss_2: 39.91562271118164
+INFO:__main__:vgg_loss_3: 28.83991813659668
+INFO:__main__:vgg_loss_4: 51.591697692871094
+INFO:__main__:vgg_loss_5: 1.6931087970733643
+INFO:__main__:validation_loss: 2482.682861328125
+INFO:__main__:global_step: 79751
+INFO:__main__:kl_loss: 98.8460693359375
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1461.235107421875
+INFO:__main__:loss: 1560.0811767578125
+INFO:__main__:vgg_gram_loss_0: 5.557261943817139
+INFO:__main__:vgg_gram_loss_1: 14.469592094421387
+INFO:__main__:vgg_gram_loss_2: 17.894466400146484
+INFO:__main__:vgg_gram_loss_3: 15.68427562713623
+INFO:__main__:vgg_gram_loss_4: 74.24887084960938
+INFO:__main__:vgg_gram_loss_5: 0.16412346065044403
+INFO:__main__:vgg_loss_0: 11.82037353515625
+INFO:__main__:vgg_loss_1: 28.306625366210938
+INFO:__main__:vgg_loss_2: 40.010498046875
+INFO:__main__:vgg_loss_3: 29.47344398498535
+INFO:__main__:vgg_loss_4: 52.8528938293457
+INFO:__main__:vgg_loss_5: 1.764564871788025
+INFO:__main__:validation_loss: 1962.7208251953125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 80001
+INFO:__main__:kl_loss: 97.75192260742188
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1553.241943359375
+INFO:__main__:loss: 1650.993896484375
+INFO:__main__:vgg_gram_loss_0: 3.591933488845825
+INFO:__main__:vgg_gram_loss_1: 17.818574905395508
+INFO:__main__:vgg_gram_loss_2: 23.086002349853516
+INFO:__main__:vgg_gram_loss_3: 17.242115020751953
+INFO:__main__:vgg_gram_loss_4: 79.01488494873047
+INFO:__main__:vgg_gram_loss_5: 0.176866814494133
+INFO:__main__:vgg_loss_0: 11.17879581451416
+INFO:__main__:vgg_loss_1: 28.167112350463867
+INFO:__main__:vgg_loss_2: 42.98210906982422
+INFO:__main__:vgg_loss_3: 31.404953002929688
+INFO:__main__:vgg_loss_4: 54.22929763793945
+INFO:__main__:vgg_loss_5: 1.7557364702224731
+INFO:__main__:validation_loss: 2135.604736328125
+INFO:__main__:global_step: 80251
+INFO:__main__:kl_loss: 103.41703796386719
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1208.22900390625
+INFO:__main__:loss: 1311.64599609375
+INFO:__main__:vgg_gram_loss_0: 3.1222074031829834
+INFO:__main__:vgg_gram_loss_1: 12.776175498962402
+INFO:__main__:vgg_gram_loss_2: 17.62489891052246
+INFO:__main__:vgg_gram_loss_3: 14.074426651000977
+INFO:__main__:vgg_gram_loss_4: 63.91703414916992
+INFO:__main__:vgg_gram_loss_5: 0.13840441405773163
+INFO:__main__:vgg_loss_0: 8.255149841308594
+INFO:__main__:vgg_loss_1: 20.59969711303711
+INFO:__main__:vgg_loss_2: 31.636394500732422
+INFO:__main__:vgg_loss_3: 23.953367233276367
+INFO:__main__:vgg_loss_4: 44.05984115600586
+INFO:__main__:vgg_loss_5: 1.4882193803787231
+INFO:__main__:validation_loss: 1988.78125
+INFO:__main__:global_step: 80501
+INFO:__main__:kl_loss: 85.78915405273438
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1344.615234375
+INFO:__main__:loss: 1430.4044189453125
+INFO:__main__:vgg_gram_loss_0: 4.585079193115234
+INFO:__main__:vgg_gram_loss_1: 12.211726188659668
+INFO:__main__:vgg_gram_loss_2: 17.756277084350586
+INFO:__main__:vgg_gram_loss_3: 14.097999572753906
+INFO:__main__:vgg_gram_loss_4: 66.73870086669922
+INFO:__main__:vgg_gram_loss_5: 0.15263496339321136
+INFO:__main__:vgg_loss_0: 11.173226356506348
+INFO:__main__:vgg_loss_1: 26.102771759033203
+INFO:__main__:vgg_loss_2: 37.94792938232422
+INFO:__main__:vgg_loss_3: 27.634408950805664
+INFO:__main__:vgg_loss_4: 48.92601776123047
+INFO:__main__:vgg_loss_5: 1.5962773561477661
+INFO:__main__:validation_loss: 2199.76025390625
+INFO:__main__:global_step: 80751
+INFO:__main__:kl_loss: 94.74278259277344
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1472.8819580078125
+INFO:__main__:loss: 1567.624755859375
+INFO:__main__:vgg_gram_loss_0: 4.285419940948486
+INFO:__main__:vgg_gram_loss_1: 15.256309509277344
+INFO:__main__:vgg_gram_loss_2: 20.529253005981445
+INFO:__main__:vgg_gram_loss_3: 16.313154220581055
+INFO:__main__:vgg_gram_loss_4: 76.03982543945312
+INFO:__main__:vgg_gram_loss_5: 0.19359265267848969
+INFO:__main__:vgg_loss_0: 10.367183685302734
+INFO:__main__:vgg_loss_1: 25.529756546020508
+INFO:__main__:vgg_loss_2: 39.88685989379883
+INFO:__main__:vgg_loss_3: 29.635881423950195
+INFO:__main__:vgg_loss_4: 54.714111328125
+INFO:__main__:vgg_loss_5: 1.8250415325164795
+INFO:__main__:validation_loss: 2002.2320556640625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 81001
+INFO:__main__:kl_loss: 98.2622299194336
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1348.048828125
+INFO:__main__:loss: 1446.31103515625
+INFO:__main__:vgg_gram_loss_0: 6.56104040145874
+INFO:__main__:vgg_gram_loss_1: 16.22667121887207
+INFO:__main__:vgg_gram_loss_2: 17.54633140563965
+INFO:__main__:vgg_gram_loss_3: 14.0277681350708
+INFO:__main__:vgg_gram_loss_4: 68.46923065185547
+INFO:__main__:vgg_gram_loss_5: 0.16178075969219208
+INFO:__main__:vgg_loss_0: 10.690400123596191
+INFO:__main__:vgg_loss_1: 25.57149314880371
+INFO:__main__:vgg_loss_2: 35.69831085205078
+INFO:__main__:vgg_loss_3: 26.339086532592773
+INFO:__main__:vgg_loss_4: 46.71260452270508
+INFO:__main__:vgg_loss_5: 1.6050379276275635
+INFO:__main__:validation_loss: 1699.84765625
+INFO:__main__:global_step: 81251
+INFO:__main__:kl_loss: 89.79828643798828
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1316.5489501953125
+INFO:__main__:loss: 1406.3472900390625
+INFO:__main__:vgg_gram_loss_0: 5.309929370880127
+INFO:__main__:vgg_gram_loss_1: 13.208151817321777
+INFO:__main__:vgg_gram_loss_2: 18.72504997253418
+INFO:__main__:vgg_gram_loss_3: 14.291874885559082
+INFO:__main__:vgg_gram_loss_4: 64.89142608642578
+INFO:__main__:vgg_gram_loss_5: 0.14290452003479004
+INFO:__main__:vgg_loss_0: 10.46673583984375
+INFO:__main__:vgg_loss_1: 25.137651443481445
+INFO:__main__:vgg_loss_2: 35.491214752197266
+INFO:__main__:vgg_loss_3: 25.999418258666992
+INFO:__main__:vgg_loss_4: 48.13096618652344
+INFO:__main__:vgg_loss_5: 1.5144715309143066
+INFO:__main__:validation_loss: 1918.27685546875
+INFO:__main__:global_step: 81501
+INFO:__main__:kl_loss: 92.11074829101562
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1847.28759765625
+INFO:__main__:loss: 1939.3983154296875
+INFO:__main__:vgg_gram_loss_0: 6.165489196777344
+INFO:__main__:vgg_gram_loss_1: 19.86362075805664
+INFO:__main__:vgg_gram_loss_2: 33.36058044433594
+INFO:__main__:vgg_gram_loss_3: 27.53851318359375
+INFO:__main__:vgg_gram_loss_4: 94.63838958740234
+INFO:__main__:vgg_gram_loss_5: 0.1914936602115631
+INFO:__main__:vgg_loss_0: 13.2963285446167
+INFO:__main__:vgg_loss_1: 33.984832763671875
+INFO:__main__:vgg_loss_2: 49.02729415893555
+INFO:__main__:vgg_loss_3: 33.766815185546875
+INFO:__main__:vgg_loss_4: 55.780391693115234
+INFO:__main__:vgg_loss_5: 1.8437576293945312
+INFO:__main__:validation_loss: 2006.922119140625
+INFO:__main__:global_step: 81751
+INFO:__main__:kl_loss: 87.86402130126953
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1526.658935546875
+INFO:__main__:loss: 1614.52294921875
+INFO:__main__:vgg_gram_loss_0: 9.031765937805176
+INFO:__main__:vgg_gram_loss_1: 17.036706924438477
+INFO:__main__:vgg_gram_loss_2: 20.1535587310791
+INFO:__main__:vgg_gram_loss_3: 16.7474422454834
+INFO:__main__:vgg_gram_loss_4: 75.70626068115234
+INFO:__main__:vgg_gram_loss_5: 0.18288543820381165
+INFO:__main__:vgg_loss_0: 11.458873748779297
+INFO:__main__:vgg_loss_1: 28.283252716064453
+INFO:__main__:vgg_loss_2: 41.22413635253906
+INFO:__main__:vgg_loss_3: 29.97891616821289
+INFO:__main__:vgg_loss_4: 53.749629974365234
+INFO:__main__:vgg_loss_5: 1.7783355712890625
+INFO:__main__:validation_loss: 1989.7784423828125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 82001
+INFO:__main__:kl_loss: 95.00089263916016
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1487.32275390625
+INFO:__main__:loss: 1582.3236083984375
+INFO:__main__:vgg_gram_loss_0: 3.9157257080078125
+INFO:__main__:vgg_gram_loss_1: 19.87759017944336
+INFO:__main__:vgg_gram_loss_2: 25.975006103515625
+INFO:__main__:vgg_gram_loss_3: 20.047096252441406
+INFO:__main__:vgg_gram_loss_4: 76.06735229492188
+INFO:__main__:vgg_gram_loss_5: 0.1750112771987915
+INFO:__main__:vgg_loss_0: 10.63952922821045
+INFO:__main__:vgg_loss_1: 27.1136417388916
+INFO:__main__:vgg_loss_2: 38.493324279785156
+INFO:__main__:vgg_loss_3: 27.182579040527344
+INFO:__main__:vgg_loss_4: 46.37953186035156
+INFO:__main__:vgg_loss_5: 1.5981570482254028
+INFO:__main__:validation_loss: 1898.049072265625
+INFO:__main__:global_step: 82251
+INFO:__main__:kl_loss: 97.98809051513672
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1274.9139404296875
+INFO:__main__:loss: 1372.9019775390625
+INFO:__main__:vgg_gram_loss_0: 4.689359664916992
+INFO:__main__:vgg_gram_loss_1: 12.45651912689209
+INFO:__main__:vgg_gram_loss_2: 17.494436264038086
+INFO:__main__:vgg_gram_loss_3: 18.9660587310791
+INFO:__main__:vgg_gram_loss_4: 72.00885009765625
+INFO:__main__:vgg_gram_loss_5: 0.13127835094928741
+INFO:__main__:vgg_loss_0: 8.620142936706543
+INFO:__main__:vgg_loss_1: 21.419042587280273
+INFO:__main__:vgg_loss_2: 31.868562698364258
+INFO:__main__:vgg_loss_3: 23.78537368774414
+INFO:__main__:vgg_loss_4: 42.12684631347656
+INFO:__main__:vgg_loss_5: 1.416313648223877
+INFO:__main__:validation_loss: 2347.246826171875
+INFO:__main__:global_step: 82501
+INFO:__main__:kl_loss: 108.59619140625
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1745.371337890625
+INFO:__main__:loss: 1853.967529296875
+INFO:__main__:vgg_gram_loss_0: 5.926742076873779
+INFO:__main__:vgg_gram_loss_1: 23.508352279663086
+INFO:__main__:vgg_gram_loss_2: 24.14362144470215
+INFO:__main__:vgg_gram_loss_3: 20.984792709350586
+INFO:__main__:vgg_gram_loss_4: 82.1069107055664
+INFO:__main__:vgg_gram_loss_5: 0.17420707643032074
+INFO:__main__:vgg_loss_0: 15.884284973144531
+INFO:__main__:vgg_loss_1: 40.44281768798828
+INFO:__main__:vgg_loss_2: 47.49895477294922
+INFO:__main__:vgg_loss_3: 32.251319885253906
+INFO:__main__:vgg_loss_4: 54.399940490722656
+INFO:__main__:vgg_loss_5: 1.7523400783538818
+INFO:__main__:validation_loss: 1879.75390625
+INFO:__main__:global_step: 82751
+INFO:__main__:kl_loss: 100.03776550292969
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1440.7900390625
+INFO:__main__:loss: 1540.8277587890625
+INFO:__main__:vgg_gram_loss_0: 5.242326259613037
+INFO:__main__:vgg_gram_loss_1: 18.2813663482666
+INFO:__main__:vgg_gram_loss_2: 19.813932418823242
+INFO:__main__:vgg_gram_loss_3: 16.2091121673584
+INFO:__main__:vgg_gram_loss_4: 72.57634735107422
+INFO:__main__:vgg_gram_loss_5: 0.16568265855312347
+INFO:__main__:vgg_loss_0: 10.24982738494873
+INFO:__main__:vgg_loss_1: 24.5814266204834
+INFO:__main__:vgg_loss_2: 38.1981315612793
+INFO:__main__:vgg_loss_3: 28.7985897064209
+INFO:__main__:vgg_loss_4: 52.30574417114258
+INFO:__main__:vgg_loss_5: 1.7355220317840576
+INFO:__main__:validation_loss: 2012.91650390625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 83001
+INFO:__main__:kl_loss: 101.6833724975586
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1602.052978515625
+INFO:__main__:loss: 1703.736328125
+INFO:__main__:vgg_gram_loss_0: 6.069923400878906
+INFO:__main__:vgg_gram_loss_1: 14.655524253845215
+INFO:__main__:vgg_gram_loss_2: 22.244489669799805
+INFO:__main__:vgg_gram_loss_3: 17.456438064575195
+INFO:__main__:vgg_gram_loss_4: 78.08736419677734
+INFO:__main__:vgg_gram_loss_5: 0.17986610531806946
+INFO:__main__:vgg_loss_0: 12.502431869506836
+INFO:__main__:vgg_loss_1: 30.994325637817383
+INFO:__main__:vgg_loss_2: 47.128055572509766
+INFO:__main__:vgg_loss_3: 32.84475326538086
+INFO:__main__:vgg_loss_4: 56.45346450805664
+INFO:__main__:vgg_loss_5: 1.7939345836639404
+INFO:__main__:validation_loss: 1939.479248046875
+INFO:__main__:global_step: 83251
+INFO:__main__:kl_loss: 97.91267395019531
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1560.0294189453125
+INFO:__main__:loss: 1657.942138671875
+INFO:__main__:vgg_gram_loss_0: 8.820950508117676
+INFO:__main__:vgg_gram_loss_1: 18.81271743774414
+INFO:__main__:vgg_gram_loss_2: 22.845487594604492
+INFO:__main__:vgg_gram_loss_3: 18.4512996673584
+INFO:__main__:vgg_gram_loss_4: 76.0999755859375
+INFO:__main__:vgg_gram_loss_5: 0.1599634885787964
+INFO:__main__:vgg_loss_0: 13.118925094604492
+INFO:__main__:vgg_loss_1: 29.68277931213379
+INFO:__main__:vgg_loss_2: 40.75514221191406
+INFO:__main__:vgg_loss_3: 29.43019676208496
+INFO:__main__:vgg_loss_4: 52.10874557495117
+INFO:__main__:vgg_loss_5: 1.7197154760360718
+INFO:__main__:validation_loss: 1903.0794677734375
+INFO:__main__:global_step: 83501
+INFO:__main__:kl_loss: 99.29400634765625
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1519.2978515625
+INFO:__main__:loss: 1618.591796875
+INFO:__main__:vgg_gram_loss_0: 7.180703639984131
+INFO:__main__:vgg_gram_loss_1: 16.98862075805664
+INFO:__main__:vgg_gram_loss_2: 19.65035057067871
+INFO:__main__:vgg_gram_loss_3: 16.420713424682617
+INFO:__main__:vgg_gram_loss_4: 75.1695327758789
+INFO:__main__:vgg_gram_loss_5: 0.17087000608444214
+INFO:__main__:vgg_loss_0: 12.835245132446289
+INFO:__main__:vgg_loss_1: 31.030221939086914
+INFO:__main__:vgg_loss_2: 41.98442077636719
+INFO:__main__:vgg_loss_3: 29.603914260864258
+INFO:__main__:vgg_loss_4: 51.14784622192383
+INFO:__main__:vgg_loss_5: 1.6771326065063477
+INFO:__main__:validation_loss: 1946.5439453125
+INFO:__main__:global_step: 83751
+INFO:__main__:kl_loss: 103.40635681152344
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1527.377197265625
+INFO:__main__:loss: 1630.7835693359375
+INFO:__main__:vgg_gram_loss_0: 8.7304048538208
+INFO:__main__:vgg_gram_loss_1: 15.662714958190918
+INFO:__main__:vgg_gram_loss_2: 21.035449981689453
+INFO:__main__:vgg_gram_loss_3: 16.605205535888672
+INFO:__main__:vgg_gram_loss_4: 76.33334350585938
+INFO:__main__:vgg_gram_loss_5: 0.16769273579120636
+INFO:__main__:vgg_loss_0: 11.964091300964355
+INFO:__main__:vgg_loss_1: 27.787216186523438
+INFO:__main__:vgg_loss_2: 41.625858306884766
+INFO:__main__:vgg_loss_3: 30.577280044555664
+INFO:__main__:vgg_loss_4: 53.25604248046875
+INFO:__main__:vgg_loss_5: 1.7301315069198608
+INFO:__main__:validation_loss: 1763.604736328125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 84001
+INFO:__main__:kl_loss: 102.80001831054688
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1866.0308837890625
+INFO:__main__:loss: 1968.8309326171875
+INFO:__main__:vgg_gram_loss_0: 4.783004283905029
+INFO:__main__:vgg_gram_loss_1: 20.34861946105957
+INFO:__main__:vgg_gram_loss_2: 27.948562622070312
+INFO:__main__:vgg_gram_loss_3: 24.024093627929688
+INFO:__main__:vgg_gram_loss_4: 97.29955291748047
+INFO:__main__:vgg_gram_loss_5: 0.20169392228126526
+INFO:__main__:vgg_loss_0: 15.449098587036133
+INFO:__main__:vgg_loss_1: 39.1492805480957
+INFO:__main__:vgg_loss_2: 49.39509201049805
+INFO:__main__:vgg_loss_3: 34.4315185546875
+INFO:__main__:vgg_loss_4: 58.24940490722656
+INFO:__main__:vgg_loss_5: 1.926268219947815
+INFO:__main__:validation_loss: 1952.7216796875
+INFO:__main__:global_step: 84251
+INFO:__main__:kl_loss: 95.65646362304688
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1656.325439453125
+INFO:__main__:loss: 1751.98193359375
+INFO:__main__:vgg_gram_loss_0: 6.038942813873291
+INFO:__main__:vgg_gram_loss_1: 17.359582901000977
+INFO:__main__:vgg_gram_loss_2: 21.96158218383789
+INFO:__main__:vgg_gram_loss_3: 19.036582946777344
+INFO:__main__:vgg_gram_loss_4: 82.67169952392578
+INFO:__main__:vgg_gram_loss_5: 0.20066487789154053
+INFO:__main__:vgg_loss_0: 13.395499229431152
+INFO:__main__:vgg_loss_1: 33.68561553955078
+INFO:__main__:vgg_loss_2: 46.19823455810547
+INFO:__main__:vgg_loss_3: 32.50307083129883
+INFO:__main__:vgg_loss_4: 56.346317291259766
+INFO:__main__:vgg_loss_5: 1.8672676086425781
+INFO:__main__:validation_loss: 1914.456298828125
+INFO:__main__:global_step: 84501
+INFO:__main__:kl_loss: 100.36820220947266
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1481.33837890625
+INFO:__main__:loss: 1581.70654296875
+INFO:__main__:vgg_gram_loss_0: 4.331390857696533
+INFO:__main__:vgg_gram_loss_1: 14.644476890563965
+INFO:__main__:vgg_gram_loss_2: 19.45265007019043
+INFO:__main__:vgg_gram_loss_3: 17.54254150390625
+INFO:__main__:vgg_gram_loss_4: 75.70055389404297
+INFO:__main__:vgg_gram_loss_5: 0.16531334817409515
+INFO:__main__:vgg_loss_0: 12.102217674255371
+INFO:__main__:vgg_loss_1: 29.424427032470703
+INFO:__main__:vgg_loss_2: 41.23120880126953
+INFO:__main__:vgg_loss_3: 29.501041412353516
+INFO:__main__:vgg_loss_4: 50.50730514526367
+INFO:__main__:vgg_loss_5: 1.6645351648330688
+INFO:__main__:validation_loss: 2069.1904296875
+INFO:__main__:global_step: 84751
+INFO:__main__:kl_loss: 94.571533203125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1551.15234375
+INFO:__main__:loss: 1645.723876953125
+INFO:__main__:vgg_gram_loss_0: 5.290731906890869
+INFO:__main__:vgg_gram_loss_1: 16.31108283996582
+INFO:__main__:vgg_gram_loss_2: 21.96208381652832
+INFO:__main__:vgg_gram_loss_3: 17.365196228027344
+INFO:__main__:vgg_gram_loss_4: 77.48499298095703
+INFO:__main__:vgg_gram_loss_5: 0.17431682348251343
+INFO:__main__:vgg_loss_0: 12.351576805114746
+INFO:__main__:vgg_loss_1: 28.668800354003906
+INFO:__main__:vgg_loss_2: 42.68034744262695
+INFO:__main__:vgg_loss_3: 31.108861923217773
+INFO:__main__:vgg_loss_4: 55.006736755371094
+INFO:__main__:vgg_loss_5: 1.8257193565368652
+INFO:__main__:validation_loss: 2112.8037109375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 85001
+INFO:__main__:kl_loss: 89.73536682128906
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1228.2298583984375
+INFO:__main__:loss: 1317.9652099609375
+INFO:__main__:vgg_gram_loss_0: 4.668101787567139
+INFO:__main__:vgg_gram_loss_1: 12.261878967285156
+INFO:__main__:vgg_gram_loss_2: 15.450517654418945
+INFO:__main__:vgg_gram_loss_3: 12.65491008758545
+INFO:__main__:vgg_gram_loss_4: 64.36504364013672
+INFO:__main__:vgg_gram_loss_5: 0.15288017690181732
+INFO:__main__:vgg_loss_0: 8.78394889831543
+INFO:__main__:vgg_loss_1: 22.08951759338379
+INFO:__main__:vgg_loss_2: 33.46101760864258
+INFO:__main__:vgg_loss_3: 24.657201766967773
+INFO:__main__:vgg_loss_4: 45.53661346435547
+INFO:__main__:vgg_loss_5: 1.5643253326416016
+INFO:__main__:validation_loss: 1937.0225830078125
+INFO:__main__:global_step: 85251
+INFO:__main__:kl_loss: 96.29811096191406
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1386.146240234375
+INFO:__main__:loss: 1482.4443359375
+INFO:__main__:vgg_gram_loss_0: 6.237278461456299
+INFO:__main__:vgg_gram_loss_1: 16.835220336914062
+INFO:__main__:vgg_gram_loss_2: 18.505664825439453
+INFO:__main__:vgg_gram_loss_3: 14.651399612426758
+INFO:__main__:vgg_gram_loss_4: 67.31338500976562
+INFO:__main__:vgg_gram_loss_5: 0.1532503068447113
+INFO:__main__:vgg_loss_0: 11.47293758392334
+INFO:__main__:vgg_loss_1: 27.777219772338867
+INFO:__main__:vgg_loss_2: 38.396507263183594
+INFO:__main__:vgg_loss_3: 27.032386779785156
+INFO:__main__:vgg_loss_4: 47.31303024291992
+INFO:__main__:vgg_loss_5: 1.5409481525421143
+INFO:__main__:validation_loss: 1697.091796875
+INFO:__main__:global_step: 85501
+INFO:__main__:kl_loss: 87.54335021972656
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1281.6160888671875
+INFO:__main__:loss: 1369.159423828125
+INFO:__main__:vgg_gram_loss_0: 3.5342705249786377
+INFO:__main__:vgg_gram_loss_1: 13.145003318786621
+INFO:__main__:vgg_gram_loss_2: 16.067468643188477
+INFO:__main__:vgg_gram_loss_3: 13.703886985778809
+INFO:__main__:vgg_gram_loss_4: 67.61222839355469
+INFO:__main__:vgg_gram_loss_5: 0.15748010575771332
+INFO:__main__:vgg_loss_0: 9.739809036254883
+INFO:__main__:vgg_loss_1: 23.411943435668945
+INFO:__main__:vgg_loss_2: 35.39277648925781
+INFO:__main__:vgg_loss_3: 25.666736602783203
+INFO:__main__:vgg_loss_4: 46.30250930786133
+INFO:__main__:vgg_loss_5: 1.5891127586364746
+INFO:__main__:validation_loss: 2179.515625
+INFO:__main__:global_step: 85751
+INFO:__main__:kl_loss: 94.78518676757812
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1712.78564453125
+INFO:__main__:loss: 1807.57080078125
+INFO:__main__:vgg_gram_loss_0: 3.6299057006835938
+INFO:__main__:vgg_gram_loss_1: 25.8809757232666
+INFO:__main__:vgg_gram_loss_2: 25.304645538330078
+INFO:__main__:vgg_gram_loss_3: 21.423974990844727
+INFO:__main__:vgg_gram_loss_4: 79.38390350341797
+INFO:__main__:vgg_gram_loss_5: 0.17424647510051727
+INFO:__main__:vgg_loss_0: 13.595664978027344
+INFO:__main__:vgg_loss_1: 34.353145599365234
+INFO:__main__:vgg_loss_2: 48.31509017944336
+INFO:__main__:vgg_loss_3: 33.50291061401367
+INFO:__main__:vgg_loss_4: 55.21525955200195
+INFO:__main__:vgg_loss_5: 1.7774025201797485
+INFO:__main__:validation_loss: 1754.9840087890625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 86001
+INFO:__main__:kl_loss: 95.00084686279297
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1531.831298828125
+INFO:__main__:loss: 1626.8321533203125
+INFO:__main__:vgg_gram_loss_0: 4.115212917327881
+INFO:__main__:vgg_gram_loss_1: 21.71470832824707
+INFO:__main__:vgg_gram_loss_2: 21.845443725585938
+INFO:__main__:vgg_gram_loss_3: 17.291040420532227
+INFO:__main__:vgg_gram_loss_4: 75.22225189208984
+INFO:__main__:vgg_gram_loss_5: 0.18292339146137238
+INFO:__main__:vgg_loss_0: 12.092474937438965
+INFO:__main__:vgg_loss_1: 30.105987548828125
+INFO:__main__:vgg_loss_2: 40.74094772338867
+INFO:__main__:vgg_loss_3: 29.764862060546875
+INFO:__main__:vgg_loss_4: 51.57970428466797
+INFO:__main__:vgg_loss_5: 1.7107014656066895
+INFO:__main__:validation_loss: 1896.9442138671875
+INFO:__main__:global_step: 86251
+INFO:__main__:kl_loss: 98.273193359375
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1558.2989501953125
+INFO:__main__:loss: 1656.5721435546875
+INFO:__main__:vgg_gram_loss_0: 5.608706951141357
+INFO:__main__:vgg_gram_loss_1: 20.91752815246582
+INFO:__main__:vgg_gram_loss_2: 22.90658187866211
+INFO:__main__:vgg_gram_loss_3: 19.253355026245117
+INFO:__main__:vgg_gram_loss_4: 73.6556625366211
+INFO:__main__:vgg_gram_loss_5: 0.17318837344646454
+INFO:__main__:vgg_loss_0: 12.03257942199707
+INFO:__main__:vgg_loss_1: 32.35426712036133
+INFO:__main__:vgg_loss_2: 42.37464904785156
+INFO:__main__:vgg_loss_3: 30.064556121826172
+INFO:__main__:vgg_loss_4: 50.6291618347168
+INFO:__main__:vgg_loss_5: 1.689544916152954
+INFO:__main__:validation_loss: 1750.8167724609375
+INFO:__main__:global_step: 86501
+INFO:__main__:kl_loss: 92.73173522949219
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1491.1016845703125
+INFO:__main__:loss: 1583.8333740234375
+INFO:__main__:vgg_gram_loss_0: 5.957441806793213
+INFO:__main__:vgg_gram_loss_1: 18.354604721069336
+INFO:__main__:vgg_gram_loss_2: 21.314176559448242
+INFO:__main__:vgg_gram_loss_3: 17.491716384887695
+INFO:__main__:vgg_gram_loss_4: 73.09115600585938
+INFO:__main__:vgg_gram_loss_5: 0.16766129434108734
+INFO:__main__:vgg_loss_0: 11.346125602722168
+INFO:__main__:vgg_loss_1: 29.41204261779785
+INFO:__main__:vgg_loss_2: 42.23638916015625
+INFO:__main__:vgg_loss_3: 28.904680252075195
+INFO:__main__:vgg_loss_4: 48.382198333740234
+INFO:__main__:vgg_loss_5: 1.562149167060852
+INFO:__main__:validation_loss: 1979.1063232421875
+INFO:__main__:global_step: 86751
+INFO:__main__:kl_loss: 93.53606414794922
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1717.9364013671875
+INFO:__main__:loss: 1811.472412109375
+INFO:__main__:vgg_gram_loss_0: 5.445728302001953
+INFO:__main__:vgg_gram_loss_1: 18.829435348510742
+INFO:__main__:vgg_gram_loss_2: 26.253692626953125
+INFO:__main__:vgg_gram_loss_3: 20.959585189819336
+INFO:__main__:vgg_gram_loss_4: 80.99195098876953
+INFO:__main__:vgg_gram_loss_5: 0.19162844121456146
+INFO:__main__:vgg_loss_0: 14.288912773132324
+INFO:__main__:vgg_loss_1: 37.115596771240234
+INFO:__main__:vgg_loss_2: 46.68117904663086
+INFO:__main__:vgg_loss_3: 33.273616790771484
+INFO:__main__:vgg_loss_4: 57.61492919921875
+INFO:__main__:vgg_loss_5: 1.9410020112991333
+INFO:__main__:validation_loss: 2241.765869140625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 87001
+INFO:__main__:kl_loss: 104.61593627929688
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1541.5565185546875
+INFO:__main__:loss: 1646.1724853515625
+INFO:__main__:vgg_gram_loss_0: 5.409865856170654
+INFO:__main__:vgg_gram_loss_1: 17.30840492248535
+INFO:__main__:vgg_gram_loss_2: 22.8099365234375
+INFO:__main__:vgg_gram_loss_3: 17.3248348236084
+INFO:__main__:vgg_gram_loss_4: 71.65416717529297
+INFO:__main__:vgg_gram_loss_5: 0.17380662262439728
+INFO:__main__:vgg_loss_0: 13.077392578125
+INFO:__main__:vgg_loss_1: 32.17558670043945
+INFO:__main__:vgg_loss_2: 44.8250617980957
+INFO:__main__:vgg_loss_3: 31.064552307128906
+INFO:__main__:vgg_loss_4: 50.850250244140625
+INFO:__main__:vgg_loss_5: 1.637457251548767
+INFO:__main__:validation_loss: 1996.9666748046875
+INFO:__main__:global_step: 87251
+INFO:__main__:kl_loss: 89.46595764160156
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1470.481201171875
+INFO:__main__:loss: 1559.9471435546875
+INFO:__main__:vgg_gram_loss_0: 6.709456920623779
+INFO:__main__:vgg_gram_loss_1: 19.351619720458984
+INFO:__main__:vgg_gram_loss_2: 22.138158798217773
+INFO:__main__:vgg_gram_loss_3: 20.97577476501465
+INFO:__main__:vgg_gram_loss_4: 72.77854919433594
+INFO:__main__:vgg_gram_loss_5: 0.16637827455997467
+INFO:__main__:vgg_loss_0: 11.251394271850586
+INFO:__main__:vgg_loss_1: 28.82857322692871
+INFO:__main__:vgg_loss_2: 38.515464782714844
+INFO:__main__:vgg_loss_3: 26.77165412902832
+INFO:__main__:vgg_loss_4: 45.11300277709961
+INFO:__main__:vgg_loss_5: 1.4962304830551147
+INFO:__main__:validation_loss: 1955.4864501953125
+INFO:__main__:global_step: 87501
+INFO:__main__:kl_loss: 92.83488464355469
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1458.609619140625
+INFO:__main__:loss: 1551.4444580078125
+INFO:__main__:vgg_gram_loss_0: 5.602681636810303
+INFO:__main__:vgg_gram_loss_1: 14.442621231079102
+INFO:__main__:vgg_gram_loss_2: 21.35629653930664
+INFO:__main__:vgg_gram_loss_3: 16.57509994506836
+INFO:__main__:vgg_gram_loss_4: 75.21454620361328
+INFO:__main__:vgg_gram_loss_5: 0.16098904609680176
+INFO:__main__:vgg_loss_0: 10.94645881652832
+INFO:__main__:vgg_loss_1: 27.643186569213867
+INFO:__main__:vgg_loss_2: 41.172218322753906
+INFO:__main__:vgg_loss_3: 28.24155616760254
+INFO:__main__:vgg_loss_4: 48.8176155090332
+INFO:__main__:vgg_loss_5: 1.5486400127410889
+INFO:__main__:validation_loss: 1766.4197998046875
+INFO:__main__:global_step: 87751
+INFO:__main__:kl_loss: 93.96218872070312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1500.3673095703125
+INFO:__main__:loss: 1594.3294677734375
+INFO:__main__:vgg_gram_loss_0: 6.3223395347595215
+INFO:__main__:vgg_gram_loss_1: 15.351176261901855
+INFO:__main__:vgg_gram_loss_2: 19.354772567749023
+INFO:__main__:vgg_gram_loss_3: 16.245092391967773
+INFO:__main__:vgg_gram_loss_4: 72.87541961669922
+INFO:__main__:vgg_gram_loss_5: 0.16141042113304138
+INFO:__main__:vgg_loss_0: 12.33423137664795
+INFO:__main__:vgg_loss_1: 30.205875396728516
+INFO:__main__:vgg_loss_2: 43.944679260253906
+INFO:__main__:vgg_loss_3: 30.77652931213379
+INFO:__main__:vgg_loss_4: 50.910518646240234
+INFO:__main__:vgg_loss_5: 1.5914095640182495
+INFO:__main__:validation_loss: 1935.1287841796875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 88001
+INFO:__main__:kl_loss: 85.048095703125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1597.5133056640625
+INFO:__main__:loss: 1682.5614013671875
+INFO:__main__:vgg_gram_loss_0: 2.612126111984253
+INFO:__main__:vgg_gram_loss_1: 18.122039794921875
+INFO:__main__:vgg_gram_loss_2: 20.531103134155273
+INFO:__main__:vgg_gram_loss_3: 16.5625
+INFO:__main__:vgg_gram_loss_4: 75.32939910888672
+INFO:__main__:vgg_gram_loss_5: 0.1899040937423706
+INFO:__main__:vgg_loss_0: 13.327157974243164
+INFO:__main__:vgg_loss_1: 33.10106658935547
+INFO:__main__:vgg_loss_2: 46.50777816772461
+INFO:__main__:vgg_loss_3: 33.21672821044922
+INFO:__main__:vgg_loss_4: 58.121768951416016
+INFO:__main__:vgg_loss_5: 1.8810677528381348
+INFO:__main__:validation_loss: 2089.902587890625
+INFO:__main__:global_step: 88251
+INFO:__main__:kl_loss: 101.83035278320312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1506.619873046875
+INFO:__main__:loss: 1608.4501953125
+INFO:__main__:vgg_gram_loss_0: 4.240794658660889
+INFO:__main__:vgg_gram_loss_1: 15.625555992126465
+INFO:__main__:vgg_gram_loss_2: 21.770673751831055
+INFO:__main__:vgg_gram_loss_3: 18.047576904296875
+INFO:__main__:vgg_gram_loss_4: 77.87972259521484
+INFO:__main__:vgg_gram_loss_5: 0.18957094848155975
+INFO:__main__:vgg_loss_0: 10.853694915771484
+INFO:__main__:vgg_loss_1: 27.552867889404297
+INFO:__main__:vgg_loss_2: 41.86859893798828
+INFO:__main__:vgg_loss_3: 30.132732391357422
+INFO:__main__:vgg_loss_4: 51.430137634277344
+INFO:__main__:vgg_loss_5: 1.732043743133545
+INFO:__main__:validation_loss: 1755.4676513671875
+INFO:__main__:global_step: 88501
+INFO:__main__:kl_loss: 100.78472137451172
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1329.21142578125
+INFO:__main__:loss: 1429.99609375
+INFO:__main__:vgg_gram_loss_0: 5.605550289154053
+INFO:__main__:vgg_gram_loss_1: 12.872700691223145
+INFO:__main__:vgg_gram_loss_2: 18.679401397705078
+INFO:__main__:vgg_gram_loss_3: 14.170722007751465
+INFO:__main__:vgg_gram_loss_4: 67.79904174804688
+INFO:__main__:vgg_gram_loss_5: 0.15615373849868774
+INFO:__main__:vgg_loss_0: 10.295143127441406
+INFO:__main__:vgg_loss_1: 24.14769744873047
+INFO:__main__:vgg_loss_2: 35.519187927246094
+INFO:__main__:vgg_loss_3: 26.667081832885742
+INFO:__main__:vgg_loss_4: 48.31919479370117
+INFO:__main__:vgg_loss_5: 1.6104235649108887
+INFO:__main__:validation_loss: 2166.90380859375
+INFO:__main__:global_step: 88751
+INFO:__main__:kl_loss: 94.95106506347656
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1810.169189453125
+INFO:__main__:loss: 1905.1202392578125
+INFO:__main__:vgg_gram_loss_0: 6.403245449066162
+INFO:__main__:vgg_gram_loss_1: 17.811805725097656
+INFO:__main__:vgg_gram_loss_2: 27.289871215820312
+INFO:__main__:vgg_gram_loss_3: 20.689159393310547
+INFO:__main__:vgg_gram_loss_4: 92.44652557373047
+INFO:__main__:vgg_gram_loss_5: 0.1994899958372116
+INFO:__main__:vgg_loss_0: 13.647820472717285
+INFO:__main__:vgg_loss_1: 33.76885986328125
+INFO:__main__:vgg_loss_2: 50.712276458740234
+INFO:__main__:vgg_loss_3: 35.57040023803711
+INFO:__main__:vgg_loss_4: 61.45115661621094
+INFO:__main__:vgg_loss_5: 2.0432472229003906
+INFO:__main__:validation_loss: 2195.940185546875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 89001
+INFO:__main__:kl_loss: 87.79652404785156
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1548.9085693359375
+INFO:__main__:loss: 1636.705078125
+INFO:__main__:vgg_gram_loss_0: 5.418597221374512
+INFO:__main__:vgg_gram_loss_1: 18.318815231323242
+INFO:__main__:vgg_gram_loss_2: 22.550067901611328
+INFO:__main__:vgg_gram_loss_3: 17.619138717651367
+INFO:__main__:vgg_gram_loss_4: 75.90799713134766
+INFO:__main__:vgg_gram_loss_5: 0.17810462415218353
+INFO:__main__:vgg_loss_0: 11.607017517089844
+INFO:__main__:vgg_loss_1: 29.64558219909668
+INFO:__main__:vgg_loss_2: 42.75007629394531
+INFO:__main__:vgg_loss_3: 30.81840705871582
+INFO:__main__:vgg_loss_4: 53.244964599609375
+INFO:__main__:vgg_loss_5: 1.722967267036438
+INFO:__main__:validation_loss: 1968.934326171875
+INFO:__main__:global_step: 89251
+INFO:__main__:kl_loss: 99.38996887207031
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1527.5535888671875
+INFO:__main__:loss: 1626.943603515625
+INFO:__main__:vgg_gram_loss_0: 6.409358501434326
+INFO:__main__:vgg_gram_loss_1: 13.839940071105957
+INFO:__main__:vgg_gram_loss_2: 19.78169059753418
+INFO:__main__:vgg_gram_loss_3: 15.48968505859375
+INFO:__main__:vgg_gram_loss_4: 75.2727279663086
+INFO:__main__:vgg_gram_loss_5: 0.19342464208602905
+INFO:__main__:vgg_loss_0: 12.747583389282227
+INFO:__main__:vgg_loss_1: 30.14181900024414
+INFO:__main__:vgg_loss_2: 44.00062561035156
+INFO:__main__:vgg_loss_3: 31.507556915283203
+INFO:__main__:vgg_loss_4: 54.32966995239258
+INFO:__main__:vgg_loss_5: 1.7966530323028564
+INFO:__main__:validation_loss: 2002.3011474609375
+INFO:__main__:global_step: 89501
+INFO:__main__:kl_loss: 79.3800048828125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1301.2054443359375
+INFO:__main__:loss: 1380.58544921875
+INFO:__main__:vgg_gram_loss_0: 6.198101043701172
+INFO:__main__:vgg_gram_loss_1: 12.816706657409668
+INFO:__main__:vgg_gram_loss_2: 16.51862907409668
+INFO:__main__:vgg_gram_loss_3: 13.665445327758789
+INFO:__main__:vgg_gram_loss_4: 66.41639709472656
+INFO:__main__:vgg_gram_loss_5: 0.16204382479190826
+INFO:__main__:vgg_loss_0: 9.489882469177246
+INFO:__main__:vgg_loss_1: 22.756256103515625
+INFO:__main__:vgg_loss_2: 35.25872039794922
+INFO:__main__:vgg_loss_3: 26.64169692993164
+INFO:__main__:vgg_loss_4: 48.70185089111328
+INFO:__main__:vgg_loss_5: 1.6153570413589478
+INFO:__main__:validation_loss: 2023.2962646484375
+INFO:__main__:global_step: 89751
+INFO:__main__:kl_loss: 94.20720672607422
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1556.41357421875
+INFO:__main__:loss: 1650.6207275390625
+INFO:__main__:vgg_gram_loss_0: 6.064948558807373
+INFO:__main__:vgg_gram_loss_1: 24.91961669921875
+INFO:__main__:vgg_gram_loss_2: 21.996809005737305
+INFO:__main__:vgg_gram_loss_3: 20.935544967651367
+INFO:__main__:vgg_gram_loss_4: 81.74285125732422
+INFO:__main__:vgg_gram_loss_5: 0.17013275623321533
+INFO:__main__:vgg_loss_0: 11.232378959655762
+INFO:__main__:vgg_loss_1: 28.35336685180664
+INFO:__main__:vgg_loss_2: 36.97306442260742
+INFO:__main__:vgg_loss_3: 27.364686965942383
+INFO:__main__:vgg_loss_4: 49.80122756958008
+INFO:__main__:vgg_loss_5: 1.7280945777893066
+INFO:__main__:validation_loss: 2082.096923828125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 90001
+INFO:__main__:kl_loss: 97.37629699707031
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1913.1427001953125
+INFO:__main__:loss: 2010.51904296875
+INFO:__main__:vgg_gram_loss_0: 6.37723970413208
+INFO:__main__:vgg_gram_loss_1: 24.330354690551758
+INFO:__main__:vgg_gram_loss_2: 36.72909164428711
+INFO:__main__:vgg_gram_loss_3: 28.11878776550293
+INFO:__main__:vgg_gram_loss_4: 91.86283111572266
+INFO:__main__:vgg_gram_loss_5: 0.1795268952846527
+INFO:__main__:vgg_loss_0: 13.628457069396973
+INFO:__main__:vgg_loss_1: 35.518577575683594
+INFO:__main__:vgg_loss_2: 52.0661735534668
+INFO:__main__:vgg_loss_3: 35.261512756347656
+INFO:__main__:vgg_loss_4: 56.80059051513672
+INFO:__main__:vgg_loss_5: 1.755385398864746
+INFO:__main__:validation_loss: 2113.16357421875
+INFO:__main__:global_step: 90251
+INFO:__main__:kl_loss: 103.89485168457031
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1303.5218505859375
+INFO:__main__:loss: 1407.416748046875
+INFO:__main__:vgg_gram_loss_0: 4.536561489105225
+INFO:__main__:vgg_gram_loss_1: 11.125533103942871
+INFO:__main__:vgg_gram_loss_2: 15.57563304901123
+INFO:__main__:vgg_gram_loss_3: 13.154977798461914
+INFO:__main__:vgg_gram_loss_4: 66.6130599975586
+INFO:__main__:vgg_gram_loss_5: 0.16151918470859528
+INFO:__main__:vgg_loss_0: 10.108838081359863
+INFO:__main__:vgg_loss_1: 24.10146713256836
+INFO:__main__:vgg_loss_2: 36.98349380493164
+INFO:__main__:vgg_loss_3: 27.460615158081055
+INFO:__main__:vgg_loss_4: 49.258060455322266
+INFO:__main__:vgg_loss_5: 1.6246066093444824
+INFO:__main__:validation_loss: 1869.824462890625
+INFO:__main__:global_step: 90501
+INFO:__main__:kl_loss: 94.542724609375
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1294.7869873046875
+INFO:__main__:loss: 1389.3297119140625
+INFO:__main__:vgg_gram_loss_0: 4.875837326049805
+INFO:__main__:vgg_gram_loss_1: 15.140176773071289
+INFO:__main__:vgg_gram_loss_2: 18.707382202148438
+INFO:__main__:vgg_gram_loss_3: 14.012394905090332
+INFO:__main__:vgg_gram_loss_4: 65.20364379882812
+INFO:__main__:vgg_gram_loss_5: 0.16046075522899628
+INFO:__main__:vgg_loss_0: 8.895411491394043
+INFO:__main__:vgg_loss_1: 23.096973419189453
+INFO:__main__:vgg_loss_2: 35.841888427734375
+INFO:__main__:vgg_loss_3: 25.902448654174805
+INFO:__main__:vgg_loss_4: 45.578739166259766
+INFO:__main__:vgg_loss_5: 1.5420160293579102
+INFO:__main__:validation_loss: 1910.3350830078125
+INFO:__main__:global_step: 90751
+INFO:__main__:kl_loss: 89.89542388916016
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1499.4200439453125
+INFO:__main__:loss: 1589.3154296875
+INFO:__main__:vgg_gram_loss_0: 5.955201148986816
+INFO:__main__:vgg_gram_loss_1: 14.562341690063477
+INFO:__main__:vgg_gram_loss_2: 18.117033004760742
+INFO:__main__:vgg_gram_loss_3: 15.038354873657227
+INFO:__main__:vgg_gram_loss_4: 74.17723083496094
+INFO:__main__:vgg_gram_loss_5: 0.1747456043958664
+INFO:__main__:vgg_loss_0: 12.216191291809082
+INFO:__main__:vgg_loss_1: 29.79570960998535
+INFO:__main__:vgg_loss_2: 43.291282653808594
+INFO:__main__:vgg_loss_3: 30.831018447875977
+INFO:__main__:vgg_loss_4: 53.94190216064453
+INFO:__main__:vgg_loss_5: 1.782986044883728
+INFO:__main__:validation_loss: 2112.77490234375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 91001
+INFO:__main__:kl_loss: 89.55815124511719
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1493.7353515625
+INFO:__main__:loss: 1583.29345703125
+INFO:__main__:vgg_gram_loss_0: 5.907973766326904
+INFO:__main__:vgg_gram_loss_1: 18.873632431030273
+INFO:__main__:vgg_gram_loss_2: 21.225566864013672
+INFO:__main__:vgg_gram_loss_3: 16.34156608581543
+INFO:__main__:vgg_gram_loss_4: 72.262451171875
+INFO:__main__:vgg_gram_loss_5: 0.17061454057693481
+INFO:__main__:vgg_loss_0: 11.40609359741211
+INFO:__main__:vgg_loss_1: 30.71196746826172
+INFO:__main__:vgg_loss_2: 42.38644790649414
+INFO:__main__:vgg_loss_3: 29.4389591217041
+INFO:__main__:vgg_loss_4: 48.474937438964844
+INFO:__main__:vgg_loss_5: 1.5468567609786987
+INFO:__main__:validation_loss: 1779.3458251953125
+INFO:__main__:global_step: 91251
+INFO:__main__:kl_loss: 104.40750122070312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1683.776611328125
+INFO:__main__:loss: 1788.18408203125
+INFO:__main__:vgg_gram_loss_0: 4.119710445404053
+INFO:__main__:vgg_gram_loss_1: 22.02119255065918
+INFO:__main__:vgg_gram_loss_2: 22.896591186523438
+INFO:__main__:vgg_gram_loss_3: 20.860820770263672
+INFO:__main__:vgg_gram_loss_4: 89.54229736328125
+INFO:__main__:vgg_gram_loss_5: 0.18235264718532562
+INFO:__main__:vgg_loss_0: 13.360998153686523
+INFO:__main__:vgg_loss_1: 33.51197814941406
+INFO:__main__:vgg_loss_2: 42.732948303222656
+INFO:__main__:vgg_loss_3: 31.197961807250977
+INFO:__main__:vgg_loss_4: 54.57076644897461
+INFO:__main__:vgg_loss_5: 1.7577097415924072
+INFO:__main__:validation_loss: 1867.725341796875
+INFO:__main__:global_step: 91501
+INFO:__main__:kl_loss: 96.52860260009766
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1225.991455078125
+INFO:__main__:loss: 1322.52001953125
+INFO:__main__:vgg_gram_loss_0: 3.574167013168335
+INFO:__main__:vgg_gram_loss_1: 15.611968994140625
+INFO:__main__:vgg_gram_loss_2: 16.45143699645996
+INFO:__main__:vgg_gram_loss_3: 14.18681526184082
+INFO:__main__:vgg_gram_loss_4: 64.59764862060547
+INFO:__main__:vgg_gram_loss_5: 0.1451445072889328
+INFO:__main__:vgg_loss_0: 8.793593406677246
+INFO:__main__:vgg_loss_1: 20.945642471313477
+INFO:__main__:vgg_loss_2: 31.84554672241211
+INFO:__main__:vgg_loss_3: 24.00324821472168
+INFO:__main__:vgg_loss_4: 43.59615707397461
+INFO:__main__:vgg_loss_5: 1.446950078010559
+INFO:__main__:validation_loss: 2403.401123046875
+INFO:__main__:global_step: 91751
+INFO:__main__:kl_loss: 98.36302185058594
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1460.9588623046875
+INFO:__main__:loss: 1559.3218994140625
+INFO:__main__:vgg_gram_loss_0: 5.331429958343506
+INFO:__main__:vgg_gram_loss_1: 16.455219268798828
+INFO:__main__:vgg_gram_loss_2: 19.17129898071289
+INFO:__main__:vgg_gram_loss_3: 15.928474426269531
+INFO:__main__:vgg_gram_loss_4: 76.09088897705078
+INFO:__main__:vgg_gram_loss_5: 0.17804428935050964
+INFO:__main__:vgg_loss_0: 10.533103942871094
+INFO:__main__:vgg_loss_1: 26.26702880859375
+INFO:__main__:vgg_loss_2: 39.90780258178711
+INFO:__main__:vgg_loss_3: 29.087141036987305
+INFO:__main__:vgg_loss_4: 51.531227111816406
+INFO:__main__:vgg_loss_5: 1.710135579109192
+INFO:__main__:validation_loss: 1860.9166259765625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 92001
+INFO:__main__:kl_loss: 106.68399047851562
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1220.4700927734375
+INFO:__main__:loss: 1327.154052734375
+INFO:__main__:vgg_gram_loss_0: 4.361277103424072
+INFO:__main__:vgg_gram_loss_1: 13.068266868591309
+INFO:__main__:vgg_gram_loss_2: 16.74493980407715
+INFO:__main__:vgg_gram_loss_3: 13.164230346679688
+INFO:__main__:vgg_gram_loss_4: 63.588558197021484
+INFO:__main__:vgg_gram_loss_5: 0.13908767700195312
+INFO:__main__:vgg_loss_0: 8.668356895446777
+INFO:__main__:vgg_loss_1: 21.730485916137695
+INFO:__main__:vgg_loss_2: 33.44631576538086
+INFO:__main__:vgg_loss_3: 24.48140525817871
+INFO:__main__:vgg_loss_4: 43.31731033325195
+INFO:__main__:vgg_loss_5: 1.3837759494781494
+INFO:__main__:validation_loss: 1851.487060546875
+INFO:__main__:global_step: 92251
+INFO:__main__:kl_loss: 84.63014221191406
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1464.8028564453125
+INFO:__main__:loss: 1549.4329833984375
+INFO:__main__:vgg_gram_loss_0: 4.079217433929443
+INFO:__main__:vgg_gram_loss_1: 15.958740234375
+INFO:__main__:vgg_gram_loss_2: 23.30150604248047
+INFO:__main__:vgg_gram_loss_3: 18.339778900146484
+INFO:__main__:vgg_gram_loss_4: 73.60639953613281
+INFO:__main__:vgg_gram_loss_5: 0.16309905052185059
+INFO:__main__:vgg_loss_0: 11.195048332214355
+INFO:__main__:vgg_loss_1: 27.027301788330078
+INFO:__main__:vgg_loss_2: 40.16344451904297
+INFO:__main__:vgg_loss_3: 28.401493072509766
+INFO:__main__:vgg_loss_4: 49.08714294433594
+INFO:__main__:vgg_loss_5: 1.637388825416565
+INFO:__main__:validation_loss: 2044.39306640625
+INFO:__main__:global_step: 92501
+INFO:__main__:kl_loss: 92.03741455078125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1344.791259765625
+INFO:__main__:loss: 1436.82861328125
+INFO:__main__:vgg_gram_loss_0: 3.1323773860931396
+INFO:__main__:vgg_gram_loss_1: 18.940420150756836
+INFO:__main__:vgg_gram_loss_2: 18.881811141967773
+INFO:__main__:vgg_gram_loss_3: 14.778685569763184
+INFO:__main__:vgg_gram_loss_4: 67.8260498046875
+INFO:__main__:vgg_gram_loss_5: 0.15912984311580658
+INFO:__main__:vgg_loss_0: 9.718703269958496
+INFO:__main__:vgg_loss_1: 24.44039535522461
+INFO:__main__:vgg_loss_2: 36.7541618347168
+INFO:__main__:vgg_loss_3: 26.59928321838379
+INFO:__main__:vgg_loss_4: 46.220882415771484
+INFO:__main__:vgg_loss_5: 1.506355881690979
+INFO:__main__:validation_loss: 1449.0714111328125
+INFO:__main__:global_step: 92751
+INFO:__main__:kl_loss: 86.67222595214844
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1256.384033203125
+INFO:__main__:loss: 1343.0562744140625
+INFO:__main__:vgg_gram_loss_0: 7.1412200927734375
+INFO:__main__:vgg_gram_loss_1: 16.490821838378906
+INFO:__main__:vgg_gram_loss_2: 18.249526977539062
+INFO:__main__:vgg_gram_loss_3: 14.088301658630371
+INFO:__main__:vgg_gram_loss_4: 62.35163116455078
+INFO:__main__:vgg_gram_loss_5: 0.15313062071800232
+INFO:__main__:vgg_loss_0: 9.92934799194336
+INFO:__main__:vgg_loss_1: 25.015121459960938
+INFO:__main__:vgg_loss_2: 32.953819274902344
+INFO:__main__:vgg_loss_3: 23.149263381958008
+INFO:__main__:vgg_loss_4: 40.36640167236328
+INFO:__main__:vgg_loss_5: 1.3882135152816772
+INFO:__main__:validation_loss: 1944.3094482421875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 93001
+INFO:__main__:kl_loss: 90.28750610351562
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1445.860595703125
+INFO:__main__:loss: 1536.1480712890625
+INFO:__main__:vgg_gram_loss_0: 3.189039468765259
+INFO:__main__:vgg_gram_loss_1: 11.682622909545898
+INFO:__main__:vgg_gram_loss_2: 16.757373809814453
+INFO:__main__:vgg_gram_loss_3: 14.558525085449219
+INFO:__main__:vgg_gram_loss_4: 73.93240356445312
+INFO:__main__:vgg_gram_loss_5: 0.17990677058696747
+INFO:__main__:vgg_loss_0: 10.843522071838379
+INFO:__main__:vgg_loss_1: 27.61995506286621
+INFO:__main__:vgg_loss_2: 42.487030029296875
+INFO:__main__:vgg_loss_3: 31.195941925048828
+INFO:__main__:vgg_loss_4: 54.88895034790039
+INFO:__main__:vgg_loss_5: 1.8368555307388306
+INFO:__main__:validation_loss: 1907.019775390625
+INFO:__main__:global_step: 93251
+INFO:__main__:kl_loss: 105.66995239257812
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1716.762939453125
+INFO:__main__:loss: 1822.432861328125
+INFO:__main__:vgg_gram_loss_0: 5.507288455963135
+INFO:__main__:vgg_gram_loss_1: 16.063644409179688
+INFO:__main__:vgg_gram_loss_2: 23.852792739868164
+INFO:__main__:vgg_gram_loss_3: 18.465246200561523
+INFO:__main__:vgg_gram_loss_4: 86.23794555664062
+INFO:__main__:vgg_gram_loss_5: 0.21492119133472443
+INFO:__main__:vgg_loss_0: 14.631569862365723
+INFO:__main__:vgg_loss_1: 31.57145118713379
+INFO:__main__:vgg_loss_2: 46.99806594848633
+INFO:__main__:vgg_loss_3: 35.329105377197266
+INFO:__main__:vgg_loss_4: 62.38677215576172
+INFO:__main__:vgg_loss_5: 2.0938076972961426
+INFO:__main__:validation_loss: 2323.74365234375
+INFO:__main__:global_step: 93501
+INFO:__main__:kl_loss: 88.48690032958984
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1581.728271484375
+INFO:__main__:loss: 1670.2152099609375
+INFO:__main__:vgg_gram_loss_0: 5.2566752433776855
+INFO:__main__:vgg_gram_loss_1: 14.151083946228027
+INFO:__main__:vgg_gram_loss_2: 20.565196990966797
+INFO:__main__:vgg_gram_loss_3: 17.462369918823242
+INFO:__main__:vgg_gram_loss_4: 77.39970397949219
+INFO:__main__:vgg_gram_loss_5: 0.17609338462352753
+INFO:__main__:vgg_loss_0: 13.479527473449707
+INFO:__main__:vgg_loss_1: 34.215274810791016
+INFO:__main__:vgg_loss_2: 46.57057189941406
+INFO:__main__:vgg_loss_3: 31.594818115234375
+INFO:__main__:vgg_loss_4: 53.76302719116211
+INFO:__main__:vgg_loss_5: 1.711329460144043
+INFO:__main__:validation_loss: 1774.981201171875
+INFO:__main__:global_step: 93751
+INFO:__main__:kl_loss: 95.56268310546875
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1587.37060546875
+INFO:__main__:loss: 1682.933349609375
+INFO:__main__:vgg_gram_loss_0: 3.1884894371032715
+INFO:__main__:vgg_gram_loss_1: 19.36784553527832
+INFO:__main__:vgg_gram_loss_2: 24.69904327392578
+INFO:__main__:vgg_gram_loss_3: 19.751630783081055
+INFO:__main__:vgg_gram_loss_4: 79.70296478271484
+INFO:__main__:vgg_gram_loss_5: 0.17828328907489777
+INFO:__main__:vgg_loss_0: 12.465038299560547
+INFO:__main__:vgg_loss_1: 32.27489471435547
+INFO:__main__:vgg_loss_2: 43.15052032470703
+INFO:__main__:vgg_loss_3: 29.957197189331055
+INFO:__main__:vgg_loss_4: 51.080204010009766
+INFO:__main__:vgg_loss_5: 1.65802001953125
+INFO:__main__:validation_loss: 2090.6181640625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 94001
+INFO:__main__:kl_loss: 95.87492370605469
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1553.4979248046875
+INFO:__main__:loss: 1649.372802734375
+INFO:__main__:vgg_gram_loss_0: 8.404108047485352
+INFO:__main__:vgg_gram_loss_1: 17.118783950805664
+INFO:__main__:vgg_gram_loss_2: 21.06252098083496
+INFO:__main__:vgg_gram_loss_3: 16.949443817138672
+INFO:__main__:vgg_gram_loss_4: 76.5724105834961
+INFO:__main__:vgg_gram_loss_5: 0.18410156667232513
+INFO:__main__:vgg_loss_0: 12.407322883605957
+INFO:__main__:vgg_loss_1: 30.708627700805664
+INFO:__main__:vgg_loss_2: 42.60052490234375
+INFO:__main__:vgg_loss_3: 30.030920028686523
+INFO:__main__:vgg_loss_4: 52.876834869384766
+INFO:__main__:vgg_loss_5: 1.7839725017547607
+INFO:__main__:validation_loss: 1884.85888671875
+INFO:__main__:global_step: 94251
+INFO:__main__:kl_loss: 92.54812622070312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1401.46337890625
+INFO:__main__:loss: 1494.011474609375
+INFO:__main__:vgg_gram_loss_0: 3.0854320526123047
+INFO:__main__:vgg_gram_loss_1: 16.0695858001709
+INFO:__main__:vgg_gram_loss_2: 20.078083038330078
+INFO:__main__:vgg_gram_loss_3: 14.960135459899902
+INFO:__main__:vgg_gram_loss_4: 67.0829086303711
+INFO:__main__:vgg_gram_loss_5: 0.1540273278951645
+INFO:__main__:vgg_loss_0: 11.4345703125
+INFO:__main__:vgg_loss_1: 28.948116302490234
+INFO:__main__:vgg_loss_2: 40.20440673828125
+INFO:__main__:vgg_loss_3: 27.985851287841797
+INFO:__main__:vgg_loss_4: 48.64845657348633
+INFO:__main__:vgg_loss_5: 1.641096830368042
+INFO:__main__:validation_loss: 1796.9822998046875
+INFO:__main__:global_step: 94501
+INFO:__main__:kl_loss: 94.13693237304688
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1449.8443603515625
+INFO:__main__:loss: 1543.9813232421875
+INFO:__main__:vgg_gram_loss_0: 5.324034214019775
+INFO:__main__:vgg_gram_loss_1: 16.135190963745117
+INFO:__main__:vgg_gram_loss_2: 19.41175651550293
+INFO:__main__:vgg_gram_loss_3: 16.938533782958984
+INFO:__main__:vgg_gram_loss_4: 75.16490936279297
+INFO:__main__:vgg_gram_loss_5: 0.15807634592056274
+INFO:__main__:vgg_loss_0: 10.632599830627441
+INFO:__main__:vgg_loss_1: 25.875459671020508
+INFO:__main__:vgg_loss_2: 38.49040985107422
+INFO:__main__:vgg_loss_3: 28.628623962402344
+INFO:__main__:vgg_loss_4: 51.511234283447266
+INFO:__main__:vgg_loss_5: 1.6980198621749878
+INFO:__main__:validation_loss: 1924.233154296875
+INFO:__main__:global_step: 94751
+INFO:__main__:kl_loss: 91.59573364257812
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1399.910888671875
+INFO:__main__:loss: 1491.506591796875
+INFO:__main__:vgg_gram_loss_0: 4.512006759643555
+INFO:__main__:vgg_gram_loss_1: 15.360969543457031
+INFO:__main__:vgg_gram_loss_2: 18.77492332458496
+INFO:__main__:vgg_gram_loss_3: 15.434577941894531
+INFO:__main__:vgg_gram_loss_4: 72.95138549804688
+INFO:__main__:vgg_gram_loss_5: 0.17269916832447052
+INFO:__main__:vgg_loss_0: 10.283180236816406
+INFO:__main__:vgg_loss_1: 24.56536865234375
+INFO:__main__:vgg_loss_2: 37.27871322631836
+INFO:__main__:vgg_loss_3: 28.04778289794922
+INFO:__main__:vgg_loss_4: 50.876163482666016
+INFO:__main__:vgg_loss_5: 1.7244006395339966
+INFO:__main__:validation_loss: 1961.497314453125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 95001
+INFO:__main__:kl_loss: 82.17642211914062
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1258.5758056640625
+INFO:__main__:loss: 1340.752197265625
+INFO:__main__:vgg_gram_loss_0: 7.235963344573975
+INFO:__main__:vgg_gram_loss_1: 13.288131713867188
+INFO:__main__:vgg_gram_loss_2: 14.8350830078125
+INFO:__main__:vgg_gram_loss_3: 12.24756145477295
+INFO:__main__:vgg_gram_loss_4: 62.040496826171875
+INFO:__main__:vgg_gram_loss_5: 0.1588146835565567
+INFO:__main__:vgg_loss_0: 9.302810668945312
+INFO:__main__:vgg_loss_1: 22.538555145263672
+INFO:__main__:vgg_loss_2: 34.76185989379883
+INFO:__main__:vgg_loss_3: 26.115793228149414
+INFO:__main__:vgg_loss_4: 47.59431838989258
+INFO:__main__:vgg_loss_5: 1.5957850217819214
+INFO:__main__:validation_loss: 2095.010498046875
+INFO:__main__:global_step: 95251
+INFO:__main__:kl_loss: 84.28132629394531
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1552.5518798828125
+INFO:__main__:loss: 1636.833251953125
+INFO:__main__:vgg_gram_loss_0: 5.086937427520752
+INFO:__main__:vgg_gram_loss_1: 20.81475257873535
+INFO:__main__:vgg_gram_loss_2: 20.502317428588867
+INFO:__main__:vgg_gram_loss_3: 16.32721519470215
+INFO:__main__:vgg_gram_loss_4: 72.61461639404297
+INFO:__main__:vgg_gram_loss_5: 0.1618853509426117
+INFO:__main__:vgg_loss_0: 12.905353546142578
+INFO:__main__:vgg_loss_1: 34.28908157348633
+INFO:__main__:vgg_loss_2: 44.21366882324219
+INFO:__main__:vgg_loss_3: 30.061050415039062
+INFO:__main__:vgg_loss_4: 51.858219146728516
+INFO:__main__:vgg_loss_5: 1.6752629280090332
+INFO:__main__:validation_loss: 1584.3267822265625
+INFO:__main__:global_step: 95501
+INFO:__main__:kl_loss: 94.78337860107422
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1478.048095703125
+INFO:__main__:loss: 1572.8314208984375
+INFO:__main__:vgg_gram_loss_0: 4.939256191253662
+INFO:__main__:vgg_gram_loss_1: 15.274670600891113
+INFO:__main__:vgg_gram_loss_2: 22.212146759033203
+INFO:__main__:vgg_gram_loss_3: 16.293222427368164
+INFO:__main__:vgg_gram_loss_4: 70.33675384521484
+INFO:__main__:vgg_gram_loss_5: 0.1675006002187729
+INFO:__main__:vgg_loss_0: 12.024515151977539
+INFO:__main__:vgg_loss_1: 30.23062515258789
+INFO:__main__:vgg_loss_2: 42.67368698120117
+INFO:__main__:vgg_loss_3: 29.679075241088867
+INFO:__main__:vgg_loss_4: 50.133636474609375
+INFO:__main__:vgg_loss_5: 1.644529938697815
+INFO:__main__:validation_loss: 1775.843994140625
+INFO:__main__:global_step: 95751
+INFO:__main__:kl_loss: 98.19869232177734
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1651.919921875
+INFO:__main__:loss: 1750.11865234375
+INFO:__main__:vgg_gram_loss_0: 4.925362586975098
+INFO:__main__:vgg_gram_loss_1: 22.744434356689453
+INFO:__main__:vgg_gram_loss_2: 21.628042221069336
+INFO:__main__:vgg_gram_loss_3: 18.437442779541016
+INFO:__main__:vgg_gram_loss_4: 82.43343353271484
+INFO:__main__:vgg_gram_loss_5: 0.1811065375804901
+INFO:__main__:vgg_loss_0: 13.052212715148926
+INFO:__main__:vgg_loss_1: 32.287330627441406
+INFO:__main__:vgg_loss_2: 43.62583541870117
+INFO:__main__:vgg_loss_3: 31.823040008544922
+INFO:__main__:vgg_loss_4: 57.32054901123047
+INFO:__main__:vgg_loss_5: 1.9251935482025146
+INFO:__main__:validation_loss: 2073.5029296875
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 96001
+INFO:__main__:kl_loss: 91.30955505371094
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1342.24267578125
+INFO:__main__:loss: 1433.55224609375
+INFO:__main__:vgg_gram_loss_0: 3.0729892253875732
+INFO:__main__:vgg_gram_loss_1: 13.904098510742188
+INFO:__main__:vgg_gram_loss_2: 16.82509994506836
+INFO:__main__:vgg_gram_loss_3: 13.945189476013184
+INFO:__main__:vgg_gram_loss_4: 69.04952239990234
+INFO:__main__:vgg_gram_loss_5: 0.16531990468502045
+INFO:__main__:vgg_loss_0: 10.11836051940918
+INFO:__main__:vgg_loss_1: 24.504222869873047
+INFO:__main__:vgg_loss_2: 37.013423919677734
+INFO:__main__:vgg_loss_3: 27.802108764648438
+INFO:__main__:vgg_loss_4: 50.36000442504883
+INFO:__main__:vgg_loss_5: 1.6882240772247314
+INFO:__main__:validation_loss: 1685.74755859375
+INFO:__main__:global_step: 96251
+INFO:__main__:kl_loss: 92.66862487792969
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1331.661376953125
+INFO:__main__:loss: 1424.3299560546875
+INFO:__main__:vgg_gram_loss_0: 5.342665195465088
+INFO:__main__:vgg_gram_loss_1: 16.640743255615234
+INFO:__main__:vgg_gram_loss_2: 20.831180572509766
+INFO:__main__:vgg_gram_loss_3: 15.615817070007324
+INFO:__main__:vgg_gram_loss_4: 64.38277435302734
+INFO:__main__:vgg_gram_loss_5: 0.16786842048168182
+INFO:__main__:vgg_loss_0: 9.960103034973145
+INFO:__main__:vgg_loss_1: 24.2823543548584
+INFO:__main__:vgg_loss_2: 36.11655807495117
+INFO:__main__:vgg_loss_3: 26.21121597290039
+INFO:__main__:vgg_loss_4: 45.203758239746094
+INFO:__main__:vgg_loss_5: 1.5772391557693481
+INFO:__main__:validation_loss: 1770.73681640625
+INFO:__main__:global_step: 96501
+INFO:__main__:kl_loss: 95.00555419921875
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1360.47998046875
+INFO:__main__:loss: 1455.485595703125
+INFO:__main__:vgg_gram_loss_0: 6.228001594543457
+INFO:__main__:vgg_gram_loss_1: 15.008150100708008
+INFO:__main__:vgg_gram_loss_2: 18.683805465698242
+INFO:__main__:vgg_gram_loss_3: 15.087199211120605
+INFO:__main__:vgg_gram_loss_4: 70.29594421386719
+INFO:__main__:vgg_gram_loss_5: 0.1775958389043808
+INFO:__main__:vgg_loss_0: 9.800933837890625
+INFO:__main__:vgg_loss_1: 24.071191787719727
+INFO:__main__:vgg_loss_2: 36.9195671081543
+INFO:__main__:vgg_loss_3: 26.72220230102539
+INFO:__main__:vgg_loss_4: 47.4777717590332
+INFO:__main__:vgg_loss_5: 1.6236779689788818
+INFO:__main__:validation_loss: 1926.9075927734375
+INFO:__main__:global_step: 96751
+INFO:__main__:kl_loss: 98.5336685180664
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1305.034423828125
+INFO:__main__:loss: 1403.568115234375
+INFO:__main__:vgg_gram_loss_0: 4.0622382164001465
+INFO:__main__:vgg_gram_loss_1: 14.667439460754395
+INFO:__main__:vgg_gram_loss_2: 17.532617568969727
+INFO:__main__:vgg_gram_loss_3: 14.533390998840332
+INFO:__main__:vgg_gram_loss_4: 66.78302001953125
+INFO:__main__:vgg_gram_loss_5: 0.1547304391860962
+INFO:__main__:vgg_loss_0: 9.376812934875488
+INFO:__main__:vgg_loss_1: 24.609041213989258
+INFO:__main__:vgg_loss_2: 36.89531326293945
+INFO:__main__:vgg_loss_3: 26.01327133178711
+INFO:__main__:vgg_loss_4: 44.89105987548828
+INFO:__main__:vgg_loss_5: 1.4879685640335083
+INFO:__main__:validation_loss: 1800.1673583984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 97001
+INFO:__main__:kl_loss: 89.10330200195312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1280.3338623046875
+INFO:__main__:loss: 1369.4371337890625
+INFO:__main__:vgg_gram_loss_0: 4.3500075340271
+INFO:__main__:vgg_gram_loss_1: 12.08788776397705
+INFO:__main__:vgg_gram_loss_2: 16.261762619018555
+INFO:__main__:vgg_gram_loss_3: 13.150627136230469
+INFO:__main__:vgg_gram_loss_4: 62.9202766418457
+INFO:__main__:vgg_gram_loss_5: 0.1583937555551529
+INFO:__main__:vgg_loss_0: 9.513590812683105
+INFO:__main__:vgg_loss_1: 24.43714714050293
+INFO:__main__:vgg_loss_2: 37.3764533996582
+INFO:__main__:vgg_loss_3: 27.102861404418945
+INFO:__main__:vgg_loss_4: 47.16448211669922
+INFO:__main__:vgg_loss_5: 1.5432795286178589
+INFO:__main__:validation_loss: 2101.328125
+INFO:__main__:global_step: 97251
+INFO:__main__:kl_loss: 89.55133819580078
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1571.9305419921875
+INFO:__main__:loss: 1661.48193359375
+INFO:__main__:vgg_gram_loss_0: 6.366440296173096
+INFO:__main__:vgg_gram_loss_1: 17.186567306518555
+INFO:__main__:vgg_gram_loss_2: 22.23857307434082
+INFO:__main__:vgg_gram_loss_3: 19.055540084838867
+INFO:__main__:vgg_gram_loss_4: 82.12162780761719
+INFO:__main__:vgg_gram_loss_5: 0.17332540452480316
+INFO:__main__:vgg_loss_0: 11.947830200195312
+INFO:__main__:vgg_loss_1: 30.00872230529785
+INFO:__main__:vgg_loss_2: 42.158203125
+INFO:__main__:vgg_loss_3: 29.955196380615234
+INFO:__main__:vgg_loss_4: 51.495567321777344
+INFO:__main__:vgg_loss_5: 1.6785327196121216
+INFO:__main__:validation_loss: 2119.100830078125
+INFO:__main__:global_step: 97501
+INFO:__main__:kl_loss: 85.68862915039062
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1413.4996337890625
+INFO:__main__:loss: 1499.188232421875
+INFO:__main__:vgg_gram_loss_0: 5.056457042694092
+INFO:__main__:vgg_gram_loss_1: 16.240156173706055
+INFO:__main__:vgg_gram_loss_2: 18.919513702392578
+INFO:__main__:vgg_gram_loss_3: 15.154683113098145
+INFO:__main__:vgg_gram_loss_4: 70.04598999023438
+INFO:__main__:vgg_gram_loss_5: 0.167255237698555
+INFO:__main__:vgg_loss_0: 11.275044441223145
+INFO:__main__:vgg_loss_1: 27.611595153808594
+INFO:__main__:vgg_loss_2: 39.640625
+INFO:__main__:vgg_loss_3: 28.003564834594727
+INFO:__main__:vgg_loss_4: 48.94350051879883
+INFO:__main__:vgg_loss_5: 1.6415457725524902
+INFO:__main__:validation_loss: 1701.56982421875
+INFO:__main__:global_step: 97751
+INFO:__main__:kl_loss: 84.13835144042969
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1457.4774169921875
+INFO:__main__:loss: 1541.61572265625
+INFO:__main__:vgg_gram_loss_0: 4.690040588378906
+INFO:__main__:vgg_gram_loss_1: 17.211193084716797
+INFO:__main__:vgg_gram_loss_2: 19.969812393188477
+INFO:__main__:vgg_gram_loss_3: 15.622014045715332
+INFO:__main__:vgg_gram_loss_4: 70.93714904785156
+INFO:__main__:vgg_gram_loss_5: 0.17449937760829926
+INFO:__main__:vgg_loss_0: 11.50278377532959
+INFO:__main__:vgg_loss_1: 28.694732666015625
+INFO:__main__:vgg_loss_2: 40.758872985839844
+INFO:__main__:vgg_loss_3: 29.34892463684082
+INFO:__main__:vgg_loss_4: 50.8730583190918
+INFO:__main__:vgg_loss_5: 1.7124176025390625
+INFO:__main__:validation_loss: 1932.9327392578125
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 98001
+INFO:__main__:kl_loss: 94.17662811279297
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1593.685546875
+INFO:__main__:loss: 1687.8621826171875
+INFO:__main__:vgg_gram_loss_0: 3.502932071685791
+INFO:__main__:vgg_gram_loss_1: 20.326452255249023
+INFO:__main__:vgg_gram_loss_2: 23.129535675048828
+INFO:__main__:vgg_gram_loss_3: 18.049726486206055
+INFO:__main__:vgg_gram_loss_4: 72.21501159667969
+INFO:__main__:vgg_gram_loss_5: 0.17129026353359222
+INFO:__main__:vgg_loss_0: 14.718999862670898
+INFO:__main__:vgg_loss_1: 36.75680160522461
+INFO:__main__:vgg_loss_2: 45.4874382019043
+INFO:__main__:vgg_loss_3: 31.008583068847656
+INFO:__main__:vgg_loss_4: 51.65411376953125
+INFO:__main__:vgg_loss_5: 1.7162220478057861
+INFO:__main__:validation_loss: 2026.9364013671875
+INFO:__main__:global_step: 98251
+INFO:__main__:kl_loss: 86.96865844726562
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1322.3330078125
+INFO:__main__:loss: 1409.3016357421875
+INFO:__main__:vgg_gram_loss_0: 4.844265937805176
+INFO:__main__:vgg_gram_loss_1: 12.612086296081543
+INFO:__main__:vgg_gram_loss_2: 17.624357223510742
+INFO:__main__:vgg_gram_loss_3: 13.71751594543457
+INFO:__main__:vgg_gram_loss_4: 66.47831726074219
+INFO:__main__:vgg_gram_loss_5: 0.15031905472278595
+INFO:__main__:vgg_loss_0: 10.065250396728516
+INFO:__main__:vgg_loss_1: 25.307106018066406
+INFO:__main__:vgg_loss_2: 37.3557243347168
+INFO:__main__:vgg_loss_3: 26.990291595458984
+INFO:__main__:vgg_loss_4: 47.74794006347656
+INFO:__main__:vgg_loss_5: 1.573432207107544
+INFO:__main__:validation_loss: 1809.330322265625
+INFO:__main__:global_step: 98501
+INFO:__main__:kl_loss: 94.30606079101562
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1458.9276123046875
+INFO:__main__:loss: 1553.233642578125
+INFO:__main__:vgg_gram_loss_0: 5.4191575050354
+INFO:__main__:vgg_gram_loss_1: 13.50175952911377
+INFO:__main__:vgg_gram_loss_2: 20.31715202331543
+INFO:__main__:vgg_gram_loss_3: 16.224205017089844
+INFO:__main__:vgg_gram_loss_4: 73.14997863769531
+INFO:__main__:vgg_gram_loss_5: 0.16184565424919128
+INFO:__main__:vgg_loss_0: 12.239108085632324
+INFO:__main__:vgg_loss_1: 30.096282958984375
+INFO:__main__:vgg_loss_2: 41.898929595947266
+INFO:__main__:vgg_loss_3: 28.560882568359375
+INFO:__main__:vgg_loss_4: 48.65317153930664
+INFO:__main__:vgg_loss_5: 1.5630378723144531
+INFO:__main__:validation_loss: 1835.910400390625
+INFO:__main__:global_step: 98751
+INFO:__main__:kl_loss: 94.64007568359375
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1659.61083984375
+INFO:__main__:loss: 1754.2509765625
+INFO:__main__:vgg_gram_loss_0: 4.645040035247803
+INFO:__main__:vgg_gram_loss_1: 24.5362606048584
+INFO:__main__:vgg_gram_loss_2: 27.10276222229004
+INFO:__main__:vgg_gram_loss_3: 21.342594146728516
+INFO:__main__:vgg_gram_loss_4: 77.76422882080078
+INFO:__main__:vgg_gram_loss_5: 0.15825985372066498
+INFO:__main__:vgg_loss_0: 13.069344520568848
+INFO:__main__:vgg_loss_1: 34.16756820678711
+INFO:__main__:vgg_loss_2: 45.76872634887695
+INFO:__main__:vgg_loss_3: 30.8070125579834
+INFO:__main__:vgg_loss_4: 50.9334602355957
+INFO:__main__:vgg_loss_5: 1.6269283294677734
+INFO:__main__:validation_loss: 1906.8941650390625
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
+INFO:__main__:global_step: 99001
+INFO:__main__:kl_loss: 90.21466064453125
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1269.030029296875
+INFO:__main__:loss: 1359.24462890625
+INFO:__main__:vgg_gram_loss_0: 3.1984193325042725
+INFO:__main__:vgg_gram_loss_1: 10.904694557189941
+INFO:__main__:vgg_gram_loss_2: 14.369215965270996
+INFO:__main__:vgg_gram_loss_3: 12.169541358947754
+INFO:__main__:vgg_gram_loss_4: 64.69881439208984
+INFO:__main__:vgg_gram_loss_5: 0.1661791205406189
+INFO:__main__:vgg_loss_0: 10.225075721740723
+INFO:__main__:vgg_loss_1: 24.334396362304688
+INFO:__main__:vgg_loss_2: 35.81004333496094
+INFO:__main__:vgg_loss_3: 26.561912536621094
+INFO:__main__:vgg_loss_4: 49.67290115356445
+INFO:__main__:vgg_loss_5: 1.6948087215423584
+INFO:__main__:validation_loss: 1952.415771484375
+INFO:__main__:global_step: 99251
+INFO:__main__:kl_loss: 108.31912231445312
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1511.6051025390625
+INFO:__main__:loss: 1619.9241943359375
+INFO:__main__:vgg_gram_loss_0: 6.157376766204834
+INFO:__main__:vgg_gram_loss_1: 18.036466598510742
+INFO:__main__:vgg_gram_loss_2: 20.51165771484375
+INFO:__main__:vgg_gram_loss_3: 17.4020938873291
+INFO:__main__:vgg_gram_loss_4: 76.51419830322266
+INFO:__main__:vgg_gram_loss_5: 0.181868776679039
+INFO:__main__:vgg_loss_0: 11.374635696411133
+INFO:__main__:vgg_loss_1: 28.961132049560547
+INFO:__main__:vgg_loss_2: 41.69327926635742
+INFO:__main__:vgg_loss_3: 29.664945602416992
+INFO:__main__:vgg_loss_4: 50.14850616455078
+INFO:__main__:vgg_loss_5: 1.674843430519104
+INFO:__main__:validation_loss: 1989.2823486328125
+INFO:__main__:global_step: 99501
+INFO:__main__:kl_loss: 99.76805114746094
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1542.787841796875
+INFO:__main__:loss: 1642.555908203125
+INFO:__main__:vgg_gram_loss_0: 7.629061222076416
+INFO:__main__:vgg_gram_loss_1: 16.972585678100586
+INFO:__main__:vgg_gram_loss_2: 24.3531436920166
+INFO:__main__:vgg_gram_loss_3: 19.431045532226562
+INFO:__main__:vgg_gram_loss_4: 77.38356018066406
+INFO:__main__:vgg_gram_loss_5: 0.1715278923511505
+INFO:__main__:vgg_loss_0: 11.751837730407715
+INFO:__main__:vgg_loss_1: 27.20474624633789
+INFO:__main__:vgg_loss_2: 40.37405014038086
+INFO:__main__:vgg_loss_3: 29.8823184967041
+INFO:__main__:vgg_loss_4: 51.6987190246582
+INFO:__main__:vgg_loss_5: 1.7049895524978638
+INFO:__main__:validation_loss: 2011.5941162109375
+INFO:__main__:global_step: 99751
+INFO:__main__:kl_loss: 93.1473388671875
+INFO:__main__:kl_weight: 1.0
+INFO:__main__:likelihood_loss: 1233.4144287109375
+INFO:__main__:loss: 1326.561767578125
+INFO:__main__:vgg_gram_loss_0: 5.318906784057617
+INFO:__main__:vgg_gram_loss_1: 12.55135440826416
+INFO:__main__:vgg_gram_loss_2: 14.961703300476074
+INFO:__main__:vgg_gram_loss_3: 12.32796573638916
+INFO:__main__:vgg_gram_loss_4: 62.49756622314453
+INFO:__main__:vgg_gram_loss_5: 0.15295223891735077
+INFO:__main__:vgg_loss_0: 9.65911865234375
+INFO:__main__:vgg_loss_1: 23.544565200805664
+INFO:__main__:vgg_loss_2: 33.62519836425781
+INFO:__main__:vgg_loss_3: 25.02433204650879
+INFO:__main__:vgg_loss_4: 45.396968841552734
+INFO:__main__:vgg_loss_5: 1.6222574710845947
+INFO:__main__:validation_loss: 1988.1361083984375
+INFO:__main__:Saved model to /net/hci-storage02/groupfolders/compvis/pesser/ma/unet/log/2017-11-06T17:04:12/checkpoints/model.ckpt
diff --git a/demo/main.py b/demo/main.py
new file mode 100644
index 00000000..99d082ed
--- /dev/null
+++ b/demo/main.py
@@ -0,0 +1,149 @@
+import tensorflow as tf
+import os
+import math
+import pickle
+import numpy as np
+import PIL
+import cv2
+import nn
+import models
+import deeploss
+
+from tqdm import trange
+from model import Model
+from batches_pg2 import plot_batch
+from get_batches import get_batches
+from utils import init_logging, process_batches
+from parser import parse_arguments
+from config import default_log_dir, config, session
+
+if __name__ == "__main__":
+    opt = parse_arguments()
+    if not os.path.exists(opt.data_index):
+        raise Exception("Invalid data index: {}".format(opt.data_index))
+
+    out_dir, logger = init_logging(opt.log_dir)
+    logger.info(opt)
+
+    if opt.mode == "train":
+        batch_size = opt.batch_size
+        img_shape = 2*[opt.spatial_size] + [3]
+        data_shape = [batch_size] + img_shape
+        init_shape = [opt.init_batches * batch_size] + img_shape
+
+        batches = get_batches(data_shape, opt.data_index, mask = opt.mask, train = True)
+        init_batches = get_batches(init_shape, opt.data_index, mask = opt.mask, train = True)
+        valid_batches = get_batches(data_shape, opt.data_index, mask = opt.mask, train = False)
+        logger.info("Number of training samples: {}".format(batches.n))
+        logger.info("Number of validation samples: {}".format(valid_batches.n))
+        if valid_batches.n == 0:
+            valid_batches = None
+
+        model = Model(opt, out_dir, logger)
+        if opt.checkpoint is not None:
+            model.restore_graph(opt.checkpoint)
+        else:
+            model.init_graph(next(init_batches))
+        model.fit(batches, valid_batches)
+
+    elif opt.mode == "test":
+        if not opt.checkpoint:
+            raise Exception("Testing requires --checkpoint")
+        batch_size = opt.batch_size
+        img_shape = 2*[opt.spatial_size] + [3]
+        data_shape = [batch_size] + img_shape
+        valid_batches = get_batches(data_shape, opt.data_index, mask = opt.mask, train = False)
+        model = Model(opt, out_dir, logger)
+        model.restore_graph(opt.checkpoint)
+
+        for i in trange(valid_batches.n // batch_size):
+            X_batch, C_batch = next(valid_batches)
+            x_gen = model.test(C_batch)
+            for k in x_gen:
+                plot_batch(x_gen[k], os.path.join(
+                    out_dir,
+                    "testing_{}_{:07}.png".format(k, i)))
+    elif opt.mode == "add_reconstructions":
+        if not opt.checkpoint:
+            raise Exception("Testing requires --checkpoint")
+        batch_size = opt.batch_size
+        img_shape = 2*[opt.spatial_size] + [3]
+        data_shape = [batch_size] + img_shape
+        batches = get_batches(data_shape, opt.data_index, mask = opt.mask,
+                train = True, return_index_id = True)
+        valid_batches = get_batches(data_shape, opt.data_index,
+                mask = opt.mask, train = False, return_index_id = True)
+        model = Model(opt, out_dir, logger)
+        model.restore_graph(opt.checkpoint)
+
+        # open index file to get image filenames and update with
+        # reconstruction data
+        with open(opt.data_index, "rb") as f:
+            index = pickle.load(f)
+        index_dir = os.path.dirname(opt.data_index)
+        index["reconstruction"] = len(index["imgs"]) * [None]
+        index["sample"] = len(index["imgs"]) * [None]
+
+        process_batches(model, index_dir, batches, batch_size, index)
+        process_batches(model, index_dir, valid_batches, batch_size, index)
+
+        # write updated index
+        with open(opt.data_index, "wb") as f:
+            pickle.dump(index, f)
+        logger.info("Wrote {}".format(opt.data_index))
+
+    elif opt.mode == "transfer":
+        if not opt.checkpoint:
+            opt.checkpoint = "log/2017-10-25T16:31:50/checkpoints/model.ckpt-100000"
+        batch_size = opt.batch_size
+        img_shape = 2*[opt.spatial_size] + [3] # outputs [256, 256, 3]
+        data_shape = [batch_size] + img_shape # outputs [8, 256, 256, 3]
+        valid_batches = get_batches(data_shape, opt.data_index,
+                mask = opt.mask, train = False)
+        model = Model(opt, out_dir, logger)
+        model.restore_graph(opt.checkpoint)
+
+        ids = ["00038", "00281", "01166", "x", "06909", "y", "07586", "07607", "z", "09874"]
+        # for step in trange(10):
+        for step in trange(10):
+            X_batch, C_batch, XN_batch, CN_batch = next(valid_batches)
+            bs = X_batch.shape[0]
+            imgs = list()
+            imgs.append(np.zeros_like(X_batch[0,...]))
+            for r in range(bs):
+                imgs.append(C_batch[r,...])
+            for i in range(bs):
+                x_infer = XN_batch[i,...]
+                c_infer = CN_batch[i,...]
+                imgs.append(X_batch[i,...])
+
+                x_infer_batch = x_infer[None,...].repeat(bs, axis = 0)
+                c_infer_batch = c_infer[None,...].repeat(bs, axis = 0)
+                c_generate_batch = C_batch
+                results = model.transfer(x_infer_batch, c_infer_batch, c_generate_batch)
+                for j in range(bs):
+                    imgs.append(results[j,...])
+            imgs = np.stack(imgs, axis = 0)
+            plot_batch(imgs, os.path.join(
+                out_dir,
+                "transfer_{}.png".format(ids[step])))
+
+    elif opt.mode == "mcmc":
+        if not opt.checkpoint:
+            raise Exception("Testing requires --checkpoint")
+        batch_size = opt.batch_size
+        img_shape = 2*[opt.spatial_size] + [3]
+        data_shape = [batch_size] + img_shape
+        valid_batches = get_batches(data_shape, opt.data_index, mask = opt.mask, train = False)
+        model = Model(opt, out_dir, logger)
+        model.restore_graph(opt.checkpoint)
+
+        for i in trange(valid_batches.n // batch_size):
+            X_batch, C_batch = next(valid_batches)
+            x_gen = model.mcmc(C_batch)
+            for k in x_gen:
+                plot_batch(x_gen[k], os.path.join(
+                    out_dir,
+                    "mcmc_{}_{:07}.png".format(k, i)))
+    else:
+        raise NotImplemented()
diff --git a/demo/model.py b/demo/model.py
new file mode 100644
index 00000000..d1cb040a
--- /dev/null
+++ b/demo/model.py
@@ -0,0 +1,416 @@
+import tensorflow as tf
+import os
+import numpy as np
+import nn
+import models
+import deeploss
+
+from tqdm import tqdm, trange
+from batches_pg2 import plot_batch, postprocess
+from get_batches import get_batches
+from config import config, session, N_BOXES
+
+class Model(object):
+    def __init__(self, opt, out_dir, logger):
+        self.batch_size = opt.batch_size
+        self.img_shape = 2*[opt.spatial_size] + [3]
+        redux = 2
+        self.imgn_shape = 2*[opt.spatial_size//(2**redux)] + [N_BOXES*3]
+        self.init_batches = opt.init_batches
+
+        self.initial_lr = opt.lr
+        self.lr_decay_begin = opt.lr_decay_begin
+        self.lr_decay_end = opt.lr_decay_end
+
+        self.out_dir = out_dir
+        self.logger = logger
+        self.log_frequency = opt.log_freq
+        self.ckpt_frequency = opt.ckpt_freq
+        self.test_frequency = opt.test_freq
+        self.checkpoint_best = False
+
+        self.dropout_p = opt.drop_prob
+
+        self.best_loss = float("inf")
+        self.checkpoint_dir = os.path.join(self.out_dir, "checkpoints")
+        os.makedirs(self.checkpoint_dir, exist_ok = True)
+
+        self.define_models()
+        self.define_graph()
+
+
+    def define_models(self):
+        n_latent_scales = 2
+        n_scales = 1 + int(np.round(np.log2(self.img_shape[0]))) - 2
+        n_filters = 32
+        redux = 2
+        self.enc_up_pass = models.make_model(
+                "enc_up", models.enc_up,
+                n_scales = n_scales - redux,
+                n_filters = n_filters*2**redux)
+        self.enc_down_pass = models.make_model(
+                "enc_down", models.enc_down,
+                n_scales = n_scales - redux,
+                n_latent_scales = n_latent_scales)
+        self.dec_up_pass = models.make_model(
+                "dec_up", models.dec_up,
+                n_scales = n_scales,
+                n_filters = n_filters)
+        self.dec_down_pass = models.make_model(
+                "dec_down", models.dec_down,
+                n_scales = n_scales,
+                n_latent_scales = n_latent_scales)
+        self.dec_params = models.make_model(
+                "dec_params", models.dec_parameters)
+
+
+    def train_forward_pass(self, x, c, xn, cn, dropout_p, init = False):
+        kwargs = {"init": init, "dropout_p": dropout_p}
+        # encoder
+        hs = self.enc_up_pass(xn, cn, **kwargs)
+        es, qs, zs_posterior = self.enc_down_pass(hs, **kwargs)
+        # decoder
+        gs = self.dec_up_pass(c, **kwargs)
+        ds, ps, zs_prior = self.dec_down_pass(gs, zs_posterior, training = True, **kwargs)
+        params = self.dec_params(ds[-1], **kwargs)
+        activations = hs + es + gs + ds
+        return params, qs, ps, activations
+
+
+    def test_forward_pass(self, c):
+        kwargs = {"init": False, "dropout_p": 0.0}
+        # decoder
+        gs = self.dec_up_pass(c, **kwargs)
+        ds, ps, zs_prior = self.dec_down_pass(gs, [], training = False, **kwargs)
+        params = self.dec_params(ds[-1], **kwargs)
+        return params
+
+
+    def transfer_pass(self, infer_x, infer_c, generate_c):
+        kwargs = {"init": False, "dropout_p": 0.0}
+        # infer latent code
+        hs = self.enc_up_pass(infer_x, infer_c, **kwargs)
+        es, qs, zs_posterior = self.enc_down_pass(hs, **kwargs)
+        zs_mean = list(qs)
+        # generate from inferred latent code and conditioning
+        gs = self.dec_up_pass(generate_c, **kwargs)
+        use_mean = True
+        if use_mean:
+            ds, ps, zs_prior = self.dec_down_pass(gs, zs_mean, training = True, **kwargs)
+        else:
+            ds, ps, zs_prior = self.dec_down_pass(gs, zs_posterior, training = True, **kwargs)
+        params = self.dec_params(ds[-1], **kwargs)
+        return params
+
+
+    def sample(self, params, **kwargs):
+        return params
+
+
+    def likelihood_loss(self, x, params):
+        return 5.0*self.vgg19.make_loss_op(x, params)
+
+
+    def define_graph(self):
+        # pretrained net for perceptual loss
+        #self.vgg19 = deeploss.JigsawFeatures(session)
+        self.vgg19 = deeploss.VGG19Features(session)
+
+        global_step = tf.Variable(0, trainable = False, name = "global_step")
+        lr = nn.make_linear_var(
+                global_step,
+                self.lr_decay_begin, self.lr_decay_end,
+                self.initial_lr, 0.0,
+                0.0, self.initial_lr)
+        kl_weight = nn.make_linear_var(
+                global_step,
+                self.lr_decay_end // 2, 3 * self.lr_decay_end // 4,
+                1e-6, 1.0,
+                1e-6, 1.0)
+        #kl_weight = tf.to_float(0.1)
+
+        # initialization
+        self.x_init = tf.placeholder(
+                tf.float32,
+                shape = [self.init_batches * self.batch_size] + self.img_shape)
+        self.c_init = tf.placeholder(
+                tf.float32,
+                shape = [self.init_batches * self.batch_size] + self.img_shape)
+        self.xn_init = tf.placeholder(
+                tf.float32,
+                shape = [self.init_batches * self.batch_size] + self.imgn_shape)
+        self.cn_init = tf.placeholder(
+                tf.float32,
+                shape = [self.init_batches * self.batch_size] + self.imgn_shape)
+        _ = self.train_forward_pass(
+                self.x_init, self.c_init,
+                self.xn_init, self.cn_init,
+                dropout_p = self.dropout_p, init = True)
+
+        # training
+        self.x = tf.placeholder(
+                tf.float32,
+                shape = [self.batch_size] + self.img_shape)
+        self.c = tf.placeholder(
+                tf.float32,
+                shape = [self.batch_size] + self.img_shape)
+        self.xn = tf.placeholder(
+                tf.float32,
+                shape = [self.batch_size] + self.imgn_shape)
+        self.cn = tf.placeholder(
+                tf.float32,
+                shape = [self.batch_size] + self.imgn_shape)
+        # compute parameters of model distribution
+        params, qs, ps, activations = self.train_forward_pass(
+                self.x, self.c,
+                self.xn, self.cn,
+                dropout_p = self.dropout_p)
+        # sample from model distribution
+        sample = self.sample(params)
+        # maximize likelihood
+        likelihood_loss = self.likelihood_loss(self.x, params)
+        kl_loss = tf.to_float(0.0)
+        for q, p in zip(qs, ps):
+            self.logger.info("Latent shape: {}".format(q.shape.as_list()))
+            kl_loss += models.latent_kl(q, p)
+        loss = likelihood_loss + kl_weight * kl_loss
+
+        # testing
+        test_forward = self.test_forward_pass(self.c)
+        test_sample = self.sample(test_forward)
+
+        # reconstruction
+        reconstruction_params, _, _, _ = self.train_forward_pass(
+                self.x, self.c,
+                self.xn, self.cn,
+                dropout_p = 0.0)
+        self.reconstruction = self.sample(reconstruction_params)
+
+        # optimization
+        self.trainable_variables = [v for v in tf.trainable_variables()
+                if not v in self.vgg19.variables]
+        optimizer = tf.train.AdamOptimizer(learning_rate = lr, beta1 = 0.5, beta2 = 0.9)
+        opt_op = optimizer.minimize(loss, var_list = self.trainable_variables)
+        with tf.control_dependencies([opt_op]):
+            self.train_op = tf.assign(global_step, global_step + 1)
+
+
+        # logging and visualization
+        self.log_ops = dict()
+        self.log_ops["global_step"] = global_step
+        self.log_ops["likelihood_loss"] = likelihood_loss
+        self.log_ops["kl_loss"] = kl_loss
+        self.log_ops["kl_weight"] = kl_weight
+        self.log_ops["loss"] = loss
+        self.img_ops = dict()
+        self.img_ops["sample"] = sample
+        self.img_ops["test_sample"] = test_sample
+        self.img_ops["x"] = self.x
+        self.img_ops["c"] = self.c
+        for i in range(N_BOXES):
+            self.img_ops["xn{}".format(i)] = self.xn[:,:,:,i*3:(i+1)*3]
+        for i, l in enumerate(self.vgg19.losses):
+            self.log_ops["vgg_loss_{}".format(i)] = l
+        for i, l in enumerate(self.vgg19.gram_losses):
+            self.log_ops["vgg_gram_loss_{}".format(i)] = l
+
+        # keep seperate train and validation summaries
+        # only training summary contains histograms
+        train_summaries = list()
+        for k, v in self.log_ops.items():
+            train_summaries.append(tf.summary.scalar(k, v))
+        self.train_summary_op = tf.summary.merge_all()
+
+        valid_summaries = list()
+        for k, v in self.log_ops.items():
+            valid_summaries.append(tf.summary.scalar(k+"_valid", v))
+        self.valid_summary_op = tf.summary.merge(valid_summaries)
+
+        # all variables for initialization
+        self.variables = [v for v in tf.global_variables()
+                if not v in self.vgg19.variables]
+
+        self.logger.info("Defined graph")
+
+
+    def init_graph(self, init_batch):
+        self.writer = tf.summary.FileWriter(
+                self.out_dir,
+                session.graph)
+        self.saver = tf.train.Saver(self.variables)
+        initializer_op = tf.variables_initializer(self.variables)
+        session.run(initializer_op, {
+            self.xn_init: init_batch[2],
+            self.cn_init: init_batch[3],
+            self.x_init: init_batch[0],
+            self.c_init: init_batch[1]})
+        self.logger.info("Initialized model from scratch")
+
+
+    def restore_graph(self, restore_path):
+        self.writer = tf.summary.FileWriter(
+                self.out_dir,
+                session.graph)
+        self.saver = tf.train.Saver(self.variables)
+        self.saver.restore(session, restore_path)
+        self.logger.info("Restored model from {}".format(restore_path))
+
+
+    def fit(self, batches, valid_batches = None):
+        session.run(tf.assign(self.log_ops["global_step"], 0))
+        start_step = self.log_ops["global_step"].eval(session)
+        self.valid_batches = valid_batches
+        for batch in trange(start_step, self.lr_decay_end):
+            X_batch, C_batch, XN_batch, CN_batch = next(batches)
+            feed_dict = {
+                    self.xn: XN_batch,
+                    self.cn: CN_batch,
+                    self.x: X_batch,
+                    self.c: C_batch}
+            fetch_dict = {"train": self.train_op}
+            if self.log_ops["global_step"].eval(session) % self.log_frequency == 0:
+                fetch_dict["log"] = self.log_ops
+                fetch_dict["img"] = self.img_ops
+                fetch_dict["summary"] = self.train_summary_op
+            result = session.run(fetch_dict, feed_dict)
+            self.log_result(result)
+
+
+    def log_result(self, result, **kwargs):
+        global_step = self.log_ops["global_step"].eval(session)
+        if "summary" in result:
+            self.writer.add_summary(result["summary"], global_step)
+            self.writer.flush()
+        if "log" in result:
+            for k in sorted(result["log"]):
+                v = result["log"][k]
+                self.logger.info("{}: {}".format(k, v))
+        if "img" in result:
+            for k, v in result["img"].items():
+                plot_batch(v, os.path.join(
+                    self.out_dir,
+                    k + "_{:07}.png".format(global_step)))
+
+            if self.valid_batches is not None:
+                # validation run
+                X_batch, C_batch, XN_batch, CN_batch = next(self.valid_batches)
+                feed_dict = {
+                        self.xn: XN_batch,
+                        self.cn: CN_batch,
+                        self.x: X_batch,
+                        self.c: C_batch}
+                fetch_dict = dict()
+                fetch_dict["imgs"] = self.img_ops
+                fetch_dict["summary"] = self.valid_summary_op
+                fetch_dict["validation_loss"] = self.log_ops["loss"]
+                result = session.run(fetch_dict, feed_dict)
+                self.writer.add_summary(result["summary"], global_step)
+                self.writer.flush()
+                # display samples
+                imgs = result["imgs"]
+                for k, v in imgs.items():
+                    plot_batch(v, os.path.join(
+                        self.out_dir,
+                        "valid_" + k + "_{:07}.png".format(global_step)))
+                # log validation loss
+                validation_loss = result["validation_loss"]
+                self.logger.info("{}: {}".format("validation_loss", validation_loss))
+                if self.checkpoint_best and validation_loss < self.best_loss:
+                    # checkpoint if validation loss improved
+                    self.logger.info("step {}: Validation loss improved from {:.4e} to {:.4e}".format(global_step, self.best_loss, validation_loss))
+                    self.best_loss = validation_loss
+                    self.make_checkpoint(global_step, prefix = "best_")
+        if global_step % self.test_frequency == 0:
+            if self.valid_batches is not None:
+                # testing
+                X_batch, C_batch, XN_batch, CN_batch = next(self.valid_batches)
+                x_gen = self.test(C_batch)
+                for k in x_gen:
+                    plot_batch(x_gen[k], os.path.join(
+                        self.out_dir,
+                        "testing_{}_{:07}.png".format(k, global_step)))
+                # transfer
+                bs = X_batch.shape[0]
+                imgs = list()
+                imgs.append(np.zeros_like(X_batch[0,...]))
+                for r in range(bs):
+                    imgs.append(C_batch[r,...])
+                for i in range(bs):
+                    x_infer = XN_batch[i,...]
+                    c_infer = CN_batch[i,...]
+                    # imgs.append(x_infer)
+                    imgs.append(X_batch[i,...])
+
+                    x_infer_batch = x_infer[None,...].repeat(bs, axis = 0)
+                    c_infer_batch = c_infer[None,...].repeat(bs, axis = 0)
+                    c_generate_batch = C_batch
+                    results = model.transfer(x_infer_batch, c_infer_batch, c_generate_batch)
+                    for j in range(bs):
+                        imgs.append(results[j,...])
+                imgs = np.stack(imgs, axis = 0)
+                plot_batch(imgs, os.path.join(
+                    out_dir,
+                    "transfer_{:07}.png".format(global_step)))
+        if global_step % self.ckpt_frequency == 0:
+            self.make_checkpoint(global_step)
+
+
+    def make_checkpoint(self, global_step, prefix = ""):
+        fname = os.path.join(self.checkpoint_dir, prefix + "model.ckpt")
+        self.saver.save(
+                session,
+                fname,
+                global_step = global_step)
+        self.logger.info("Saved model to {}".format(fname))
+
+
+    def test(self, c_batch):
+        results = dict()
+        results["cond"] = c_batch
+        sample = session.run(self.img_ops["test_sample"],
+            {self.c: c_batch})
+        results["test_sample"] = sample
+        return results
+
+
+    def mcmc(self, c_batch, n_iters = 10):
+        results = dict()
+        results["cond"] = c_batch
+        sample = session.run(
+            self.img_ops["test_sample"], {self.c: c_batch})
+        results["sample_{}".format(0)] = sample
+        for i in range(n_iters - 1):
+            sample = session.run(
+                    self.img_ops["sample"], {
+                        self.x: sample,
+                        self.c: c_batch})
+            results["sample_{:03}".format(i+1)] = sample
+        return results
+
+
+    def reconstruct(self, x_batch, c_batch):
+        return session.run(
+                self.reconstruction,
+                {self.x: x_batch, self.c: c_batch})
+
+
+    def transfer(self, x_encode, c_encode, c_decode):
+        initialized = getattr(self, "_init_transfer", False)
+        if not initialized:
+            # transfer
+            self.c_generator = tf.placeholder(
+                    tf.float32,
+                    shape = [self.batch_size] + self.img_shape)
+            infer_x = self.xn
+            infer_c = self.cn
+            generate_c = self.c_generator
+            transfer_params = self.transfer_pass(infer_x, infer_c, generate_c)
+            self.transfer_mean_sample = self.sample(transfer_params)
+            self._init_transfer = True
+
+        return session.run(
+                self.transfer_mean_sample, {
+                    self.xn: x_encode,
+                    self.cn: c_encode,
+                    self.c_generator: c_decode})
+
diff --git a/demo/models.py b/demo/models.py
new file mode 100644
index 00000000..b0023b89
--- /dev/null
+++ b/demo/models.py
@@ -0,0 +1,355 @@
+import tensorflow as tf
+import numpy as np
+from tensorflow.contrib.framework.python.ops import arg_scope
+import nn
+import math
+
+
+def model_arg_scope(**kwargs):
+    """Create new counter and apply arg scope to all arg scoped nn
+    operations."""
+    counters = {}
+    return arg_scope(
+            [nn.conv2d, nn.deconv2d, nn.residual_block, nn.dense, nn.activate],
+            counters = counters, **kwargs)
+
+
+def make_model(name, template, **kwargs):
+    """Create model with fixed kwargs."""
+    run = lambda *args, **kw: template(*args, **dict((k, v) for kws in (kw, kwargs) for k, v in kws.items()))
+    return tf.make_template(name, run, unique_name_ = name)
+
+
+def dec_up(
+        c, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        h = nn.nin(c, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        return hs
+
+
+def dec_down(
+        gs, zs_posterior, training, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu",
+        n_latent_scales = 2):
+    assert n_residual_blocks % 2 == 0
+    gs = list(gs)
+    zs_posterior = list(zs_posterior)
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = [] # hidden units
+        ps = [] # priors
+        zs = [] # prior samples
+        # prepare input
+        n_filters = gs[-1].shape.as_list()[-1]
+        h = nn.nin(gs[-1], n_filters)
+        for l in range(n_scales):
+            # level module
+            ## hidden units
+            for i in range(n_residual_blocks // 2):
+                h = nn.residual_block(h, gs.pop())
+                hs.append(h)
+            if l < n_latent_scales:
+                ## prior
+                spatial_shape = h.shape.as_list()[1]
+                n_h_channels = h.shape.as_list()[-1]
+                if spatial_shape == 1:
+                    ### no spatial correlations
+                    p = latent_parameters(h)
+                    ps.append(p)
+                    z_prior = latent_sample(p)
+                    zs.append(z_prior)
+                else:
+                    ### four autoregressively modeled groups
+                    if training:
+                        z_posterior_groups = nn.split_groups(zs_posterior[0])
+                    p_groups = []
+                    z_groups = []
+                    p_features = tf.space_to_depth(nn.residual_block(h), 2)
+                    for i in range(4):
+                        p_group = latent_parameters(p_features, num_filters = n_h_channels)
+                        p_groups.append(p_group)
+                        z_group = latent_sample(p_group)
+                        z_groups.append(z_group)
+                        # ar feedback sampled from
+                        if training:
+                            feedback = z_posterior_groups.pop(0)
+                        else:
+                            feedback = z_group
+                        # prepare input for next group
+                        if i + 1 < 4:
+                            p_features = nn.residual_block(p_features, feedback)
+                    if training:
+                        assert not z_posterior_groups
+                    # complete prior parameters
+                    p = nn.merge_groups(p_groups)
+                    ps.append(p)
+                    # complete prior sample
+                    z_prior = nn.merge_groups(z_groups)
+                    zs.append(z_prior)
+                ## vae feedback sampled from
+                if training:
+                    ## posterior
+                    z = zs_posterior.pop(0)
+                else:
+                    ## prior
+                    z = z_prior
+                for i in range(n_residual_blocks // 2):
+                    n_h_channels = h.shape.as_list()[-1]
+                    h = tf.concat([h, z], axis = -1)
+                    h = nn.nin(h, n_h_channels)
+                    h = nn.residual_block(h, gs.pop())
+                    hs.append(h)
+            else:
+                for i in range(n_residual_blocks // 2):
+                    h = nn.residual_block(h, gs.pop())
+                    hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = gs[-1].shape.as_list()[-1]
+                h = nn.upsample(h, n_filters)
+
+        assert not gs
+        if training:
+            assert not zs_posterior
+
+        return hs, ps, zs
+
+
+def encoder(
+        x, n_out, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        xc = x
+        h = nn.nin(xc, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        h = nn.nin(h, n_out)
+        hs.append(h)
+        return hs
+
+
+def feature_encoder(
+        x, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        xc = x
+        h = nn.nin(xc, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        return hs
+
+
+def cfn(
+        x, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        xc = x
+        h = nn.nin(xc, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        h_shape = h.shape.as_list()
+        h = tf.reshape(h, [h_shape[0],1,1,h_shape[1]*h_shape[2]*h_shape[3]])
+        h = nn.nin(h, 2*max_filters)
+        hs.append(h)
+        return hs
+
+
+def cfn_features(
+        x, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        xc = x
+        h = nn.nin(xc, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        return hs
+
+
+def classifier(
+        x, n_out, init = False, dropout_p = 0.5,
+        activation = "elu"):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        x_shape = x.shape.as_list()#tf.shape(x)
+        h = tf.reshape(x, [x_shape[0], 1, 1, x_shape[1]*x_shape[2]*x_shape[3]])
+        h = nn.activate(h)
+        h = nn.nin(h, 1024)
+        h = nn.activate(h)
+        h = nn.nin(h, n_out)
+        h = tf.reshape(h, [x_shape[0], n_out])
+        return h
+
+
+def enc_up(
+        x, c, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu", n_filters = 64, max_filters = 128):
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = []
+        # prepare input
+        #xc = tf.concat([x,c], axis = -1)
+        xc = x
+        h = nn.nin(xc, n_filters)
+        for l in range(n_scales):
+            # level module
+            for i in range(n_residual_blocks):
+                h = nn.residual_block(h)
+                hs.append(h)
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = min(2*n_filters, max_filters)
+                h = nn.downsample(h, n_filters)
+        return hs
+
+
+def enc_down(
+        gs, init = False, dropout_p = 0.5,
+        n_scales = 1, n_residual_blocks = 2, activation = "elu",
+        n_latent_scales = 2):
+    assert n_residual_blocks % 2 == 0
+    gs = list(gs)
+    with model_arg_scope(
+            init = init, dropout_p = dropout_p, activation = activation):
+        # outputs
+        hs = [] # hidden units
+        qs = [] # posteriors
+        zs = [] # samples from posterior
+        # prepare input
+        n_filters = gs[-1].shape.as_list()[-1]
+        h = nn.nin(gs[-1], n_filters)
+        for l in range(n_scales):
+            # level module
+            ## hidden units
+            for i in range(n_residual_blocks // 2):
+                h = nn.residual_block(h, gs.pop())
+                hs.append(h)
+            if l < n_latent_scales:
+                ## posterior parameters
+                q = latent_parameters(h)
+                qs.append(q)
+                ## posterior sample
+                z = latent_sample(q)
+                zs.append(z)
+                ## sample feedback
+                for i in range(n_residual_blocks // 2):
+                    gz = tf.concat([gs.pop(), z], axis = -1)
+                    h = nn.residual_block(h, gz)
+                    hs.append(h)
+            else:
+                """ no need to go down any further
+                for i in range(n_residual_blocks // 2):
+                    h = nn.residual_block(h, gs.pop())
+                    hs.append(h)
+                """
+                break
+            # prepare input to next level
+            if l + 1 < n_scales:
+                n_filters = gs[-1].shape.as_list()[-1]
+                h = nn.upsample(h, n_filters)
+
+        #assert not gs # not true anymore since we break out of the loop
+
+        return hs, qs, zs
+
+
+# Distributions
+
+
+def dec_parameters(
+        h, init = False, **kwargs):
+    with model_arg_scope(init = init):
+        num_filters = 3
+        return nn.conv2d(h, num_filters)
+
+
+def latent_parameters(
+        h, init = False, **kwargs):
+    num_filters = kwargs.get("num_filters", h.shape.as_list()[-1])
+    return nn.conv2d(h, num_filters)
+
+
+def logvarvar(u):
+    cutoff = tf.to_float(-5)
+    logvar = tf.maximum(cutoff, u)
+    var = tf.exp(logvar)
+    return logvar, var
+
+
+def latent_sample(p):
+    mean = p
+    stddev = 1.0
+    eps = tf.random_normal(mean.shape, mean = 0.0, stddev = 1.0)
+    return mean + stddev * eps
+
+
+def latent_kl(q, p):
+    mean1 = q
+    mean2 = p
+
+    kl = 0.5 * tf.square(mean2 - mean1)
+    kl = tf.reduce_sum(kl, axis = [1,2,3])
+    kl = tf.reduce_mean(kl)
+    return kl
diff --git a/demo/nn.py b/demo/nn.py
new file mode 100644
index 00000000..51f1f3c2
--- /dev/null
+++ b/demo/nn.py
@@ -0,0 +1,521 @@
+"""
+modified from pixelcnn++
+Various tensorflow utilities
+"""
+
+import numpy as np
+import tensorflow as tf
+from tensorflow.contrib.framework.python.ops import add_arg_scope
+
+
+def int_shape(x):
+    return x.shape.as_list()
+
+
+def concat_elu(x):
+    """ like concatenated ReLU (http://arxiv.org/abs/1603.05201), but then with ELU """
+    axis = len(x.get_shape()) - 1
+    return tf.nn.elu(tf.concat([x, -x], axis))
+
+
+def log_sum_exp(x):
+    """ numerically stable log_sum_exp implementation that prevents overflow """
+    axis = len(x.get_shape()) - 1
+    m = tf.reduce_max(x, axis)
+    m2 = tf.reduce_max(x, axis, keep_dims=True)
+    return m + tf.log(tf.reduce_sum(tf.exp(x - m2), axis))
+
+
+def log_prob_from_logits(x):
+    """ numerically stable log_softmax implementation that prevents overflow """
+    axis = len(x.get_shape()) - 1
+    m = tf.reduce_max(x, axis, keep_dims=True)
+    return x - m - tf.log(tf.reduce_sum(tf.exp(x - m), axis, keep_dims=True))
+
+
+def ce_loss(x, l):
+    x = (x + 1.0) / 2.0
+    x = tf.clip_by_value(255 * x, 0, 255)
+    x = tf.cast(x, tf.int32)
+    reconst_cost = tf.reduce_mean(
+            tf.nn.sparse_softmax_cross_entropy_with_logits(
+                logits=tf.reshape(l, [-1, 256]),
+                labels=tf.reshape(x, [-1])
+                )
+            )
+    return reconst_cost
+
+
+def ce_sample(logits, temp = 1.0):
+    temp = tf.maximum(tf.convert_to_tensor(1e-5), tf.convert_to_tensor(temp))
+    noise = tf.random_uniform(logits.shape, minval = 1e-5, maxval = 1.0 - 1e-5)
+    pixels = tf.argmax(logits / temp - tf.log(-tf.log(noise)), 4)
+    pixels = tf.cast(pixels, tf.float32) / 127.5 - 1.0
+    return pixels
+
+
+def discretized_mix_logistic_loss(x, l, sum_all=True):
+    """ log-likelihood for mixture of discretized logistics, assumes the data has been rescaled to [-1,1] interval """
+    xs = int_shape(
+        x)  # true image (i.e. labels) to regress to, e.g. (B,32,32,3)
+    ls = int_shape(l)  # predicted distribution, e.g. (B,32,32,100)
+    # here and below: unpacking the params of the mixture of logistics
+    nr_mix = int(ls[-1] / 10)
+    logit_probs = l[:, :, :, :nr_mix]
+    l = tf.reshape(l[:, :, :, nr_mix:], xs + [nr_mix * 3])
+    means = l[:, :, :, :, :nr_mix]
+    log_scales = tf.maximum(l[:, :, :, :, nr_mix:2 * nr_mix], -2.)
+    coeffs = tf.nn.tanh(l[:, :, :, :, 2 * nr_mix:3 * nr_mix])
+    # here and below: getting the means and adjusting them based on preceding
+    # sub-pixels
+    x = tf.reshape(x, xs + [1]) + tf.zeros(xs + [nr_mix])
+    m2 = tf.reshape(means[:, :, :, 1, :] + coeffs[:, :, :, 0, :]
+                    * x[:, :, :, 0, :], [xs[0], xs[1], xs[2], 1, nr_mix])
+    m3 = tf.reshape(means[:, :, :, 2, :] + coeffs[:, :, :, 1, :] * x[:, :, :, 0, :] +
+                    coeffs[:, :, :, 2, :] * x[:, :, :, 1, :], [xs[0], xs[1], xs[2], 1, nr_mix])
+    means = tf.concat([tf.reshape(means[:, :, :, 0, :], [
+                      xs[0], xs[1], xs[2], 1, nr_mix]), m2, m3], 3)
+    centered_x = x - means
+    inv_stdv = tf.exp(-log_scales)
+    plus_in = inv_stdv * (centered_x + 1. / 255.)
+    cdf_plus = tf.nn.sigmoid(plus_in)
+    min_in = inv_stdv * (centered_x - 1. / 255.)
+    cdf_min = tf.nn.sigmoid(min_in)
+    # log probability for edge case of 0 (before scaling)
+    log_cdf_plus = plus_in - tf.nn.softplus(plus_in)
+    # log probability for edge case of 255 (before scaling)
+    log_one_minus_cdf_min = -tf.nn.softplus(min_in)
+    cdf_delta = cdf_plus - cdf_min  # probability for all other cases
+    mid_in = inv_stdv * centered_x
+    # log probability in the center of the bin, to be used in extreme cases
+    # (not actually used in our code)
+    log_pdf_mid = -mid_in - log_scales - 2. * tf.nn.softplus(-mid_in)
+
+    # now select the right output: left edge case, right edge case, normal
+    # case, extremely low prob case (doesn't actually happen for us)
+
+    # this is what we are really doing, but using the robust version below for extreme cases in other applications and to avoid NaN issue with tf.select()
+    # log_probs = tf.select(x < -0.999, log_cdf_plus, tf.select(x > 0.999, log_one_minus_cdf_min, tf.log(cdf_delta)))
+
+    # robust version, that still works if probabilities are below 1e-5 (which never happens in our code)
+    # tensorflow backpropagates through tf.select() by multiplying with zero instead of selecting: this requires use to use some ugly tricks to avoid potential NaNs
+    # the 1e-12 in tf.maximum(cdf_delta, 1e-12) is never actually used as output, it's purely there to get around the tf.select() gradient issue
+    # if the probability on a sub-pixel is below 1e-5, we use an approximation
+    # based on the assumption that the log-density is constant in the bin of
+    # the observed sub-pixel value
+    log_probs = tf.where(x < -0.999, log_cdf_plus, tf.where(x > 0.999, log_one_minus_cdf_min,
+                                                            tf.where(cdf_delta > 1e-5, tf.log(tf.maximum(cdf_delta, 1e-12)), log_pdf_mid - np.log(127.5))))
+
+    log_probs = tf.reduce_sum(log_probs, 3) + log_prob_from_logits(logit_probs)
+    return -tf.reduce_mean(tf.reduce_sum(log_sum_exp(log_probs), [1, 2]))
+
+
+def sample_from_discretized_mix_logistic(l, nr_mix, temp1 = 1.0, temp2 = 1.0, mean = False):
+    if mean:
+        temp2 = 0.0
+    ls = int_shape(l)
+    xs = ls[:-1] + [3]
+    # unpack parameters
+    logit_probs = l[:, :, :, :nr_mix]
+    l = tf.reshape(l[:, :, :, nr_mix:], xs + [nr_mix * 3])
+    # sample mixture indicator from softmax
+    if not mean:
+        if temp1 < 1e-5:
+            sel = tf.one_hot(tf.argmax(logit_probs, 3), depth=nr_mix, dtype=tf.float32)
+        else:
+            sel = tf.one_hot(tf.argmax(logit_probs/temp1 - tf.log(-tf.log(tf.random_uniform(
+                logit_probs.get_shape(), minval=1e-5, maxval=1. - 1e-5))), 3), depth=nr_mix, dtype=tf.float32)
+    else:
+        sel = tf.nn.softmax(logit_probs)
+    sel = tf.reshape(sel, xs[:-1] + [1, nr_mix])
+    # select logistic parameters
+    means = tf.reduce_sum(l[:, :, :, :, :nr_mix] * sel, 4)
+    log_scales = tf.maximum(tf.reduce_sum(
+        l[:, :, :, :, nr_mix:2 * nr_mix] * sel, 4), -2.)
+    coeffs = tf.reduce_sum(tf.nn.tanh(
+        l[:, :, :, :, 2 * nr_mix:3 * nr_mix]) * sel, 4)
+    # sample from logistic & clip to interval
+    # we don't actually round to the nearest 8bit value when sampling
+    u = tf.random_uniform(means.get_shape(), minval=1e-5, maxval=1. - 1e-5)
+    x = means + temp2 * tf.exp(log_scales) * (tf.log(u) - tf.log(1. - u))
+    x0 = tf.minimum(tf.maximum(x[:, :, :, 0], -1.), 1.)
+    x1 = tf.minimum(tf.maximum(
+        x[:, :, :, 1] + coeffs[:, :, :, 0] * x0, -1.), 1.)
+    x2 = tf.minimum(tf.maximum(
+        x[:, :, :, 2] + coeffs[:, :, :, 1] * x0 + coeffs[:, :, :, 2] * x1, -1.), 1.)
+    return tf.concat([tf.reshape(x0, xs[:-1] + [1]), tf.reshape(x1, xs[:-1] + [1]), tf.reshape(x2, xs[:-1] + [1])], 3)
+
+
+''' layers containing trainable variables. '''
+
+def get_name(layer_name, counters):
+    ''' utlity for keeping track of layer names '''
+    if not layer_name in counters:
+        counters[layer_name] = 0
+    name = layer_name + '_' + str(counters[layer_name])
+    counters[layer_name] += 1
+    return name
+
+
+@add_arg_scope
+def dense(x, num_units, init_scale=1., counters={}, init=False, **kwargs):
+    ''' fully connected layer '''
+    name = get_name('dense', counters)
+    with tf.variable_scope(name):
+        if init:
+            xs = x.shape.as_list()
+            # data based initialization of parameters
+            V = tf.get_variable('V', [xs[1], num_units], tf.float32, tf.random_normal_initializer(0, 0.05))
+            V_norm = tf.nn.l2_normalize(V.initialized_value(), [0])
+            x_init = tf.matmul(x, V_norm)
+            m_init, v_init = tf.nn.moments(x_init, [0])
+            scale_init = init_scale / tf.sqrt(v_init + 1e-10)
+            g = tf.get_variable('g', dtype=tf.float32, initializer=scale_init)
+            b = tf.get_variable('b', dtype=tf.float32, initializer=-m_init * scale_init)
+            x_init = tf.reshape(scale_init, [1, num_units]) * (x_init - tf.reshape(m_init, [1, num_units]))
+
+            return x_init
+        else:
+            V = tf.get_variable("V")
+            g = tf.get_variable("g")
+            b = tf.get_variable("b")
+            with tf.control_dependencies([tf.assert_variables_initialized([V, g, b])]):
+                # use weight normalization (Salimans & Kingma, 2016)
+                x = tf.matmul(x, V)
+                scaler = g / tf.sqrt(tf.reduce_sum(tf.square(V), [0]))
+                x = tf.reshape(scaler, [1, num_units]) * x + tf.reshape(b, [1, num_units])
+
+                return x
+
+
+@add_arg_scope
+def conv2d(x, num_filters, filter_size=[3, 3], stride=[1, 1], pad='SAME', init_scale=1., counters={}, init=False, **kwargs):
+    ''' convolutional layer '''
+    num_filters = int(num_filters)
+    strides = [1] + stride + [1]
+    name = get_name('conv2d', counters)
+    with tf.variable_scope(name):
+        if init:
+            xs = x.shape.as_list()
+            # data based initialization of parameters
+            V = tf.get_variable('V', filter_size + [xs[-1], num_filters],
+                                tf.float32, tf.random_normal_initializer(0, 0.05))
+            V_norm = tf.nn.l2_normalize(V.initialized_value(), [0, 1, 2])
+            x_init = tf.nn.conv2d(x, V_norm, strides, pad)
+            m_init, v_init = tf.nn.moments(x_init, [0, 1, 2])
+            scale_init = init_scale / tf.sqrt(v_init + 1e-8)
+            g = tf.get_variable('g', dtype=tf.float32, initializer = scale_init)
+            b = tf.get_variable('b', dtype=tf.float32, initializer = -m_init * scale_init)
+            x_init = tf.reshape(scale_init, [1, 1, 1, num_filters]) * (x_init - tf.reshape(m_init, [1, 1, 1, num_filters]))
+
+            return x_init
+        else:
+            V = tf.get_variable("V")
+            g = tf.get_variable("g")
+            b = tf.get_variable("b")
+            with tf.control_dependencies([tf.assert_variables_initialized([V, g, b])]):
+                # use weight normalization (Salimans & Kingma, 2016)
+                W = tf.reshape(g, [1, 1, 1, num_filters]) * tf.nn.l2_normalize(V, [0, 1, 2])
+
+                # calculate convolutional layer output
+                x = tf.nn.bias_add(tf.nn.conv2d(x, W, strides, pad), b)
+
+                return x
+
+
+@add_arg_scope
+def deconv2d(x, num_filters, filter_size=[3, 3], stride=[1, 1], pad='SAME', init_scale=1., counters={}, init=False, **kwargs):
+    ''' transposed convolutional layer '''
+    num_filters = int(num_filters)
+    name = get_name('deconv2d', counters)
+    xs = int_shape(x)
+    strides = [1] + stride + [1]
+    if pad == 'SAME':
+        target_shape = [xs[0], xs[1] * stride[0],
+                        xs[2] * stride[1], num_filters]
+    else:
+        target_shape = [xs[0], xs[1] * stride[0] + filter_size[0] -
+                        1, xs[2] * stride[1] + filter_size[1] - 1, num_filters]
+    with tf.variable_scope(name):
+        if init:
+            # data based initialization of parameters
+            V = tf.get_variable('V', filter_size + [num_filters, xs[-1]], tf.float32, tf.random_normal_initializer(0, 0.05))
+            V_norm = tf.nn.l2_normalize(V.initialized_value(), [0, 1, 3])
+            x_init = tf.nn.conv2d_transpose(x, V_norm, target_shape, strides, padding=pad)
+            m_init, v_init = tf.nn.moments(x_init, [0, 1, 2])
+            scale_init = init_scale / tf.sqrt(v_init + 1e-8)
+            g = tf.get_variable('g', dtype=tf.float32, initializer=scale_init)
+            b = tf.get_variable('b', dtype=tf.float32, initializer=-m_init * scale_init)
+            x_init = tf.reshape(scale_init, [1, 1, 1, num_filters]) * (x_init - tf.reshape(m_init, [1, 1, 1, num_filters]))
+
+            return x_init
+        else:
+            V = tf.get_variable("V")
+            g = tf.get_variable("g")
+            b = tf.get_variable("b")
+            with tf.control_dependencies([tf.assert_variables_initialized([V, g, b])]):
+                # use weight normalization (Salimans & Kingma, 2016)
+                W = tf.reshape(g, [1, 1, num_filters, 1]) * tf.nn.l2_normalize(V, [0, 1, 3])
+
+                # calculate convolutional layer output
+                x = tf.nn.conv2d_transpose(x, W, target_shape, strides, padding=pad)
+                x = tf.nn.bias_add(x, b)
+
+                return x
+
+
+@add_arg_scope
+def activate(x, activation, **kwargs):
+    if activation == None:
+        return x
+    elif activation == "elu":
+        return tf.nn.elu(x)
+    else:
+        raise NotImplemented(activation)
+
+
+''' meta-layer consisting of multiple base layers '''
+
+def nin(x, num_units):
+    """ a network in network layer (1x1 CONV) """
+    s = int_shape(x)
+    x = tf.reshape(x, [np.prod(s[:-1]), s[-1]])
+    x = dense(x, num_units)
+    return tf.reshape(x, s[:-1] + [num_units])
+
+
+def downsample(x, num_units):
+    return conv2d(x, num_units, stride = [2, 2])
+
+
+def upsample(x, num_units, method = "subpixel"):
+    if method == "conv_transposed":
+        return deconv2d(x, num_units, stride = [2, 2])
+    elif method == "subpixel":
+        x = conv2d(x, 4*num_units)
+        x = tf.depth_to_space(x, 2)
+        return x
+
+
+@add_arg_scope
+def residual_block(x, a = None, conv=conv2d, init=False, dropout_p=0.0, gated = False, **kwargs):
+    """Slight variation of original."""
+    xs = int_shape(x)
+    num_filters = xs[-1]
+
+    residual = x
+    if a is not None:
+        a = nin(activate(a), num_filters)
+        residual = tf.concat([residual, a], axis = -1)
+    residual = activate(residual)
+    residual = tf.nn.dropout(residual, keep_prob = 1.0 - dropout_p)
+    residual = conv(residual, num_filters)
+    if gated:
+        residual = activate(residual)
+        residual = tf.nn.dropout(residual, keep_prob = 1.0 - dropout_p)
+        residual = conv(residual, 2*num_filters)
+        a, b = tf.split(residual, 2, 3)
+        residual = a * tf.nn.sigmoid(b)
+
+    return x + residual
+
+
+''' utilities for shifting the image around, efficient alternative to masking convolutions '''
+
+def down_shift(x):
+    xs = int_shape(x)
+    return tf.concat([tf.zeros([xs[0], 1, xs[2], xs[3]]), x[:, :xs[1] - 1, :, :]], 1)
+
+
+def right_shift(x):
+    xs = int_shape(x)
+    return tf.concat([tf.zeros([xs[0], xs[1], 1, xs[3]]), x[:, :, :xs[2] - 1, :]], 2)
+
+
+@add_arg_scope
+def down_shifted_conv2d(x, num_filters, filter_size=[2, 3], stride=[1, 1], **kwargs):
+    x = tf.pad(x, [[0, 0], [filter_size[0] - 1, 0],
+                   [int((filter_size[1] - 1) / 2), int((filter_size[1] - 1) / 2)], [0, 0]])
+    return conv2d(x, num_filters, filter_size=filter_size, pad='VALID', stride=stride, **kwargs)
+
+
+@add_arg_scope
+def down_shifted_deconv2d(x, num_filters, filter_size=[2, 3], stride=[1, 1], **kwargs):
+    x = deconv2d(x, num_filters, filter_size=filter_size,
+                 pad='VALID', stride=stride, **kwargs)
+    xs = int_shape(x)
+    return x[:, :(xs[1] - filter_size[0] + 1), int((filter_size[1] - 1) / 2):(xs[2] - int((filter_size[1] - 1) / 2)), :]
+
+
+@add_arg_scope
+def down_right_shifted_conv2d(x, num_filters, filter_size=[2, 2], stride=[1, 1], **kwargs):
+    x = tf.pad(x, [[0, 0], [filter_size[0] - 1, 0],
+                   [filter_size[1] - 1, 0], [0, 0]])
+    return conv2d(x, num_filters, filter_size=filter_size, pad='VALID', stride=stride, **kwargs)
+
+
+@add_arg_scope
+def down_right_shifted_deconv2d(x, num_filters, filter_size=[2, 2], stride=[1, 1], **kwargs):
+    x = deconv2d(x, num_filters, filter_size=filter_size,
+                 pad='VALID', stride=stride, **kwargs)
+    xs = int_shape(x)
+    return x[:, :(xs[1] - filter_size[0] + 1):, :(xs[2] - filter_size[1] + 1), :]
+
+
+################################################################ random tf stuff
+
+
+def make_linear_var(
+        step,
+        start, end,
+        start_value, end_value,
+        clip_min = 0.0, clip_max = 1.0):
+    """linear from (a, alpha) to (b, beta), i.e.
+    (beta - alpha)/(b - a) * (x - a) + alpha"""
+    linear = (
+            (end_value - start_value) /
+            (end - start) *
+            (tf.cast(step, tf.float32) - start) + start_value)
+    return tf.clip_by_value(linear, clip_min, clip_max)
+
+"""Simple approximation of 2d gaussian kernel."""
+k = np.float32([1,4,6,4,1])
+k = np.outer(k, k)
+# normalize and extend to three independent input and output channels
+kernel = k[:,:,None,None]/k.sum()*np.eye(3, dtype=np.float32)
+def tf_gaussian_subsample(x):
+    return tf.nn.conv2d(
+            input = x,
+            filter = kernel,
+            strides = [1, 2, 2, 1],
+            padding = "SAME")
+
+
+# stride 2 subsampling
+nnkernel = np.eye(3)[None,None,:,:]
+def tf_subsample(x):
+    return tf.nn.conv2d(
+            input = x,
+            filter = nnkernel,
+            strides = [1, 2, 2, 1],
+            padding = "SAME")
+
+
+# replace
+tf_downsample = tf_gaussian_subsample
+
+
+def tf_pyramid(x, ps, p2 = None):
+    """Pyramid of x, coarse to fine"""
+    nd = round(math.log(ps, 2))
+    assert 2**nd == ps
+    xs = x.get_shape().as_list()
+    assert len(xs) == 4
+    b, h, w, c = xs
+    assert h == w
+    if p2 is None:
+        p2 = round(math.log(h,ps))
+        assert ps**p2 == h, "{}, {}, {}".format(ps, p2, h)
+    pyramid = [x]
+    for i in range(p2):
+        p = pyramid[-1]
+        for j in range(nd):
+            p = tf_downsample(p)
+        pyramid.append(p)
+    return list(reversed(pyramid))
+
+
+def tf_concat_coarse_cond(coarse, cond):
+    groups = tf.concat([coarse] + 3* [tf.zeros_like(coarse)], axis = 3)
+    coarse_up = tf.depth_to_space(groups, 2)
+    return tf.concat([coarse_up, cond], axis = 3)
+
+
+def np_concat_coarse_cond(coarse, cond):
+    groups = np.concatenate([coarse] + 3* [np.zeros_like(coarse)], axis = 3)
+    coarse_up = np_depth_to_space(groups, 2)
+    return np.concatenate([coarse_up, cond], axis = 3)
+
+
+smoothing1d = np.float32([1,2,1])
+difference1d = np.float32([1,0,-1])
+sobelx = np.outer(smoothing1d, difference1d)
+sobely = np.transpose(sobelx)
+# one dim for number of input channels
+sobelx = sobelx[:,:,None]
+sobely = sobely[:,:,None]
+# stack along new dim for output channels
+sobel = np.stack([sobelx, sobely], axis = -1)
+
+fdx = np.zeros([3,3])
+fdx[1,:] = difference1d
+fdx = fdx[:,:,None]
+
+fdy = np.zeros([3,3])
+fdy[:,1] = difference1d
+fdy = fdy[:,:,None]
+fd = np.stack([fdx, fdy], axis = -1)
+def tf_img_grad(x, use_sobel = True):
+    """Sobel approximation of gradient."""
+    gray = tf.reduce_mean(x, axis = -1, keep_dims = True)
+    if use_sobel:
+        filter_ = sobel
+    else:
+        filter_ = fd
+    grad = tf.nn.conv2d(
+            input = gray,
+            filter = filter_,
+            strides = 4*[1],
+            padding = "SAME")
+    return grad
+
+
+def tf_grad_loss(x, y):
+    """Mean squared L2 difference of gradients."""
+    gx = tf_img_grad(x)
+    gy = tf_img_grad(y)
+    return tf.reduce_mean(tf.contrib.layers.flatten(tf.square(gx - gy)))
+
+
+def tf_grad_mag(x):
+    """Pointwise L2 norm of gradient."""
+    gx = tf_img_grad(x)
+    return tf.sqrt(tf.reduce_sum(tf.square(gx), axis = -1, keep_dims = True))
+
+
+def tv_loss(x):
+    h = 1.0 / x.shape.as_list()[1]
+    g = tf_img_grad(x, use_sobel = False)
+    hgl1 = h * tf.sqrt(
+            tf.reduce_sum(
+                tf.square(g),
+                axis = 3))
+    return tf.reduce_mean(
+            tf.reduce_sum(
+                hgl1,
+                axis = [1,2]))
+
+
+
+def likelihood_loss(target, tail_decoding, loss):
+    if loss == "l2":
+        rec_loss = tf.reduce_mean(tf.contrib.layers.flatten(
+            tf.square(target - tail_decoding)))
+    elif loss == "l1":
+        rec_loss = tf.reduce_mean(tf.contrib.layers.flatten(
+            tf.abs(target - tail_decoding)))
+    elif loss == "h1":
+        rec_loss = tf.reduce_mean(tf.contrib.layers.flatten(
+            tf.square(target - tail_decoding)))
+        rec_loss += tf_grad_loss(target, tail_decoding)
+    else:
+        raise NotImplemented("Unknown loss function: {}".format(loss))
+    return rec_loss
+
+
+def split_groups(x, bs = 2):
+    return tf.split(tf.space_to_depth(x, bs), bs**2, axis = 3)
+
+
+def merge_groups(xs, bs = 2):
+    return tf.depth_to_space(tf.concat(xs, axis = 3), bs)
diff --git a/demo/parser.py b/demo/parser.py
new file mode 100644
index 00000000..a99fb949
--- /dev/null
+++ b/demo/parser.py
@@ -0,0 +1,25 @@
+import argparse
+from config import default_log_dir
+
+def parse_arguments():
+    parser = argparse.ArgumentParser()
+    parser.add_argument("--data_index", required = True, help = "path to training or testing data index")
+    parser.add_argument("--mode", default = "train",
+            choices=["train", "test", "mcmc", "add_reconstructions", "transfer"])
+    parser.add_argument("--log_dir", default = default_log_dir, help = "path to log into")
+    parser.add_argument("--batch_size", default = 8, type = int, help = "batch size")
+    parser.add_argument("--init_batches", default = 4, type = int, help = "number of batches for initialization")
+    parser.add_argument("--checkpoint", help = "path to checkpoint to restore")
+    parser.add_argument("--spatial_size", default = 256, type = int, help = "spatial size to resize images to")
+    parser.add_argument("--lr", default = 1e-3, type = float, help = "initial learning rate")
+    parser.add_argument("--lr_decay_begin", default = 1000, type = int, help = "steps after which to begin linear lr decay")
+    parser.add_argument("--lr_decay_end", default = 100000, type = int, help = "step at which lr is zero, i.e. number of training steps")
+    parser.add_argument("--log_freq", default = 250, type = int, help = "frequency to log")
+    parser.add_argument("--ckpt_freq", default = 1000, type = int, help = "frequency to checkpoint")
+    parser.add_argument("--test_freq", default = 1000, type = int, help = "frequency to test")
+    parser.add_argument("--drop_prob", default = 0.1, type = float, help = "Dropout probability")
+    parser.add_argument("--mask", dest = "mask", action = "store_true", help = "Use masked data")
+    parser.add_argument("--no-mask", dest = "mask", action = "store_false", help = "Do not use mask")
+    parser.set_defaults(mask = True)
+    return parser.parse_args()
+
diff --git a/demo/requirements.txt b/demo/requirements.txt
new file mode 100644
index 00000000..5411ad05
--- /dev/null
+++ b/demo/requirements.txt
@@ -0,0 +1,7 @@
+tensorflow==1.2.1
+numpy==1.14.5
+opencv-python==3.4.3.18
+Pillow==5.2.0
+tqdm==4.26.0
+PyYAML==3.13
+h5py==2.8.0
diff --git a/demo/run.sh b/demo/run.sh
new file mode 100755
index 00000000..ccb6d602
--- /dev/null
+++ b/demo/run.sh
@@ -0,0 +1,19 @@
+#! /usr/bin/env bash
+
+python main.py  --mode "transfer" \
+                --data_index ./datasets/deepfashion/index.p \
+                --log_dir ./log \
+                --batch_size 8 \
+                --init_batches 4 \
+                --checkpoint ./checkpoints/model.ckpt-100000 \
+                --spatial_size 256 \
+                --lr 0.001 \
+                --lr_decay_begin 1000 \
+                --lr_decay_end 100000 \
+                --log_freq 250 \
+                --ckpt_freq 1000 \
+                --test_freq 1000 \
+                --drop_prob 0.1 \
+                --mask
+                # --no-mask
+
diff --git a/demo/test.py b/demo/test.py
new file mode 100644
index 00000000..04b8389b
--- /dev/null
+++ b/demo/test.py
@@ -0,0 +1,55 @@
+from PIL import Image, ImageDraw, ImageFont
+import numpy as np
+
+def draw_rotated_text(image, angle, xy, text, font):
+    """ Draw text at an angle into an image, takes the same arguments
+        as Image.text() except for:
+
+    :param image: Image to write text into
+    :param angle: Angle to write text at
+    """
+    width, height = image.size
+    max_dim = max(width, height)
+    mask_size = (max_dim * 2, max_dim * 2)
+    mask = Image.new('L', mask_size, 0)
+    draw = ImageDraw.Draw(mask)
+    draw.text((max_dim, max_dim), text, 255, font=font)
+    bigger_mask = mask.resize((max_dim*8, max_dim*8),
+                                resample=Image.BICUBIC)
+    rotated_mask = bigger_mask.rotate(angle).resize(
+        mask_size, resample=Image.LANCZOS)
+    mask_xy = (max_dim - xy[0], max_dim - xy[1])
+    b_box = mask_xy + (mask_xy[0] + width, mask_xy[1] + height)
+    mask = rotated_mask.crop(b_box)
+    color_image = Image.new('RGBA', image.size, (255,255,255))
+    image.paste(color_image, mask)
+
+def draw_middle_line(image, angle, xy):
+    width, height = image.size
+    max_dim = max(width, height)
+    mask_size = (max_dim * 2, max_dim * 2)
+    mask = Image.new('L', mask_size, 0)
+    draw = ImageDraw.Draw(mask)
+    draw.line([(0,0), (width, height)], fill=255, width=7)
+    bigger_mask = mask.resize((max_dim*8, max_dim*8),
+                                resample=Image.BICUBIC)
+    rotated_mask = bigger_mask.rotate(angle).resize(
+        mask_size, resample=Image.LANCZOS)
+    mask_xy = (max_dim - xy[0], max_dim - xy[1])
+    b_box = mask_xy + (mask_xy[0] + width, mask_xy[1] + height)
+    mask = rotated_mask.crop(b_box)
+    color_image = Image.new('RGBA', image.size, (255,255,255))
+    image.paste(color_image, mask)
+
+img = np.zeros((256, 256, 3), dtype=np.uint8)
+img = Image.fromarray(img)
+font = ImageFont.truetype("./arial.ttf", 32)
+w,h = img.size
+
+draw_middle_line(img, -45, (100,100))
+draw_rotated_text(img, -45, (150, 25), 'shape', font=font)
+draw_rotated_text(img, -45, (30, 100), 'appearance', font=font)
+
+img.show()
+
+
diff --git a/demo/timer.py b/demo/timer.py
new file mode 100644
index 00000000..538e1b37
--- /dev/null
+++ b/demo/timer.py
@@ -0,0 +1,17 @@
+import time
+
+class Timer(object):
+    def __init__(self):
+        self.tick()
+
+
+    def tick(self):
+        self.start_time = time.time()
+
+
+    def tock(self):
+        self.end_time = time.time()
+        time_since_tick = self.end_time - self.start_time
+        self.tick()
+        return time_since_tick
+
diff --git a/demo/utils.py b/demo/utils.py
new file mode 100644
index 00000000..e75a4abe
--- /dev/null
+++ b/demo/utils.py
@@ -0,0 +1,54 @@
+import os
+import datetime
+import shutil
+import glob
+import logging
+import math
+import PIL
+
+from tqdm import trange
+from batches_pg2 import postprocess
+
+
+def init_logging(out_base_dir):
+    # get unique output directory based on current time
+    os.makedirs(out_base_dir, exist_ok = True)
+    now = datetime.datetime.now().strftime("%Y-%m-%dT%H:%M:%S")
+    out_dir = os.path.join(out_base_dir, now)
+    os.makedirs(out_dir, exist_ok = False)
+    # copy source code to logging dir to have an idea what the run was about
+    this_file = os.path.realpath(__file__)
+    assert(this_file.endswith(".py"))
+    shutil.copy(this_file, out_dir)
+    # copy all py files to logging dir
+    src_dir = os.path.dirname(this_file)
+    py_files = glob.glob(os.path.join(src_dir, "*.py"))
+    for py_file in py_files:
+        shutil.copy(py_file, out_dir)
+    # init logging
+    logging.basicConfig(filename = os.path.join(out_dir, 'log.txt'))
+    logger = logging.getLogger(__name__)
+    logger.setLevel(logging.DEBUG)
+    return out_dir, logger
+
+
+def process_batches(model, index_dir, batches, batch_size, index):
+    for i in trange(math.ceil(batches.n / batch_size)):
+        X_batch, C_batch, I_batch = next(batches)
+        # reconstructions
+        R_batch = model.reconstruct(X_batch, C_batch)
+        R_batch = postprocess(R_batch) # to uint8 for saving
+        # samples from pose
+        S_batch = model.test(C_batch)["test_sample"]
+        S_batch = postprocess(S_batch) # to uint8 for saving
+        for batch_i, i in enumerate(I_batch):
+            original_fname = index["imgs"][i]
+            reconstr_fname = original_fname.rsplit(".", 1)[0] + "_reconstruction.png"
+            reconstr_path = os.path.join(index_dir, reconstr_fname)
+            sample_fname = original_fname.rsplit(".", 1)[0] + "_sample.png"
+            sample_path = os.path.join(index_dir, sample_fname)
+            index["reconstruction"][i] = reconstr_path
+            index["sample"][i] = sample_path
+            PIL.Image.fromarray(R_batch[batch_i,...]).save(reconstr_path)
+            PIL.Image.fromarray(S_batch[batch_i,...]).save(sample_path)
+