Benchmarking-Initiative · dweindl · Nov 5, 2025
@@ -38,6 +38,8 @@ jobs:
         run: src/python/build_site.py
       - name: Create overview
         run: bmp-create-overview --html-file=_site/index.html
+      - name: Create reports
+        run: python3 vis/generate_petab_reports.py --output-dir=_site/reports
       - name: Setup Pages
         uses: actions/configure-pages@v5
       - name: Upload artifact

@@ -28,7 +28,7 @@ dependencies = [
 
 [project.optional-dependencies]
 dev = ["pre-commit", "pytest", "ruff"]
-site = ["bokeh>=3.7.3"]
+site = ["bokeh>=3.7.3", "petab[vis]", "Jinja2>=3.0.3"]
 
 [project.scripts]
 bmp-petablint = "benchmark_models_petab.check_petablint:main"

@@ -0,0 +1,97 @@
+import argparse
+import base64
+import io
+from pathlib import Path
+
+import matplotlib.pyplot as plt
+import pandas as pd
+import petab.v1
+import petab.v1.visualize.plotter
+from jinja2 import Environment, FileSystemLoader
+from petab.visualize import plot_problem
+
+import benchmark_models_petab as bmp
+
+# plot simulation line without markers
+petab.v1.visualize.plotter.simulation_line_kwargs["marker"] = ""
+# plot measurements with markers only
+petab.v1.visualize.plotter.measurement_line_kwargs["linestyle"] = "None"
+
+
+def _plot_problem(problem: petab.v1.Problem, sim_df: pd.DataFrame) -> str:
+    """Plot measurement points and simulation line for one observable and return base64 png."""
+    plot_problem(problem, simulations_df=sim_df)
+    fig = plt.gcf()
+
+    buf = io.BytesIO()
+    fig.savefig(buf, format="png", dpi=150)
+    plt.close(fig)
+    buf.seek(0)
+    img_b64 = base64.b64encode(buf.read()).decode("ascii")
+    return img_b64
+
+
+def generate_report_for_model(problem_id: str, template_env: Environment, out_dir: Path):
+    problem = bmp.get_problem(problem_id)
+    if problem is None:
+        return
+
+    sim_df = bmp.get_simulation_df(problem_id)
+
+    images: list[dict[str, str]] = []
+    img = _plot_problem(problem, sim_df)
+    images.append({"id": problem_id, "img": img})
+
+    template = template_env.get_template("problem_report.html.jinja")
+    rendered = template.render(model_name=problem_id, images=images)
+
+    out_dir.mkdir(parents=True, exist_ok=True)
+    out_path = out_dir / f"{problem_id}.html"
+    with open(out_path, "w", encoding="utf-8") as fh:
+        fh.write(rendered)
+    print(f"Wrote {out_path}")
+
+
+def generate_index(problem_id_list: list[str], template_env: Environment, out_dir: Path):
+    """Render an index page with links to per-problem reports."""
+    template = template_env.get_template("index.html.jinja")
+    rendered = template.render(models=problem_id_list, count=len(problem_id_list))
+
+    out_dir.mkdir(parents=True, exist_ok=True)
+    out_path = out_dir / "index.html"
+    with open(out_path, "w", encoding="utf-8") as fh:
+        fh.write(rendered)
+    print(f"Wrote {out_path}")
+
+
+def main(output_dir: Path = None):
+    if output_dir is None:
+        output_dir = Path(__file__).parent / "reports"
+    templates_dir = Path(__file__).parent / "templates"
+
+    env = Environment(loader=FileSystemLoader(templates_dir), autoescape=True)
+    problem_id_list = list(bmp.MODELS)
+
+    for problem_id in problem_id_list:
+        try:
+            generate_report_for_model(problem_id, env, output_dir)
+        except Exception as e:
+            print(f"Skipping {problem_id} due to error: {e}")
+
+    try:
+        generate_index(problem_id_list, env, output_dir)
+    except Exception as e:
+        print(f"Failed to write index: {e}")
+
+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(description="Generate PETab problem reports")
+    parser.add_argument(
+        "--output-dir",
+        "-o",
+        type=Path,
+        default=None,
+        help="Directory to write reports to (default: vis/reports)",
+    )
+    args = parser.parse_args()
+
+    main(args.output_dir)
@@ -0,0 +1,26 @@
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8">
+    <title>PEtab problems overview</title>
+    <style>
+      body { font-family: Arial, sans-serif; margin: 20px; }
+      ul { list-style: none; padding: 0; }
+      li { margin: 6px 0; }
+      a { text-decoration: none; color: #0366d6; }
+      a:hover { text-decoration: underline; }
+    </style>
+  </head>
+  <body>
+    <h1>PEtab problems ({{ count }})</h1>
+    {% if models %}
+      <ul>
+        {% for m in models %}
+          <li><a href="{{ m }}.html">{{ m }}</a></li>
+        {% endfor %}
+      </ul>
+    {% else %}
+      <p>No problems found.</p>
+    {% endif %}
+  </body>
+</html>
@@ -0,0 +1,38 @@
+<!-- html -->
+<!doctype html>
+<html lang="en">
+  <head>
+    <meta charset="utf-8">
+    <title>{{ model_name }} - PEtab report</title>
+    <style>
+      body { font-family: Arial, sans-serif; margin: 20px; }
+      .figure { margin-bottom: 24px; }
+      img { max-width: 100%; height: auto; border: 1px solid #ccc; }
+      .top-nav { margin-bottom: 12px; }
+      .top-nav a { text-decoration: none; color: #0366d6; margin-right: 12px; }
+      .top-nav a:hover { text-decoration: underline; }
+    </style>
+  </head>
+  <body>
+    <div class="top-nav">
+      <a href="index.html">&larr; Back to index</a>
+      <a href="https://github.com/Benchmarking-Initiative/Benchmark-Models-PEtab/tree/master/Benchmark-Models/{{ model_name|urlencode }}" target="_blank" rel="noopener noreferrer">
+        View PEtab files on GitHub
+      </a>
+    </div>
+
+    <h1>{{ model_name }}</h1>
+    {% if images %}
+      {% for img in images %}
+        <div class="figure">
+          {% if images|length > 1 %}
+            <h3>{{ img.id }}</h3>
+          {% endif %}
+          <img src="data:image/png;base64,{{ img.img }}" alt="{{ img.id }}">
+        </div>
+      {% endfor %}
+    {% else %}
+      <p>No observables or data available for this problem.</p>
+    {% endif %}
+  </body>
+</html>