fix tests/test_models.py::test_model_dirs_have_metadata - AssertionError: Missing metadata file: /home/runner/work/matbench-discovery/matbench-discovery/models/alignn_ff/metadata.yml

assert False

Author: janosh

.github/workflows/gh-pages.yml

@@ -13,7 +13,6 @@ jobs:
     with:
       python-version: "3.11"
       working-directory: site
-      install-cmd: npm install --force
       pre-build: |
         pip install lazydocs
         # lazydocs needs package deps to be installed

scripts/model_figs/cumulative_metrics.py

@@ -64,6 +64,10 @@
         font=dict(size=14),
     )
     fig.update_traces(line=dict(width=3))
+    fig.layout.legend.update(
+        orientation="h", yanchor="bottom", y=1.1, xanchor="center", x=0.5
+    )
+
     for trace in fig.data:
         # show only the N best models by default
         if trace.name in df_metrics.T.sort_values("F1").index[:-6]:
@@ -109,4 +113,4 @@
 # %%
 img_name = f"cumulative-{'-'.join(metrics).lower()}"
 save_fig(fig, f"{FIGS}/{img_name}.svelte")
-save_fig(fig, f"{PDF_FIGS}/{img_name}.pdf", width=900, height=330)
+save_fig(fig, f"{PDF_FIGS}/{img_name}.pdf", width=900, height=400)

site/src/figs/cumulative-precision-recall.svelte

@@ -1645,6 +1645,60 @@ references:
   URL: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8489092/
   volume: '22'
 
+- id: lan_adsorbml_2023
+  abstract: >-
+    Computational catalysis is playing an increasingly significant role in the
+    design of catalysts across a wide range of applications. A common task for
+    many computational methods is the need to accurately compute the minimum
+    binding energy - the adsorption energy - for an adsorbate and a catalyst
+    surface of interest. Traditionally, the identification of low energy
+    adsorbate-surface configurations relies on heuristic methods and researcher
+    intuition. As the desire to perform high-throughput screening increases, it
+    becomes challenging to use heuristics and intuition alone. In this paper, we
+    demonstrate machine learning potentials can be leveraged to identify low
+    energy adsorbate-surface configurations more accurately and efficiently. Our
+    algorithm provides a spectrum of trade-offs between accuracy and efficiency,
+    with one balanced option finding the lowest energy configuration, within a
+    0.1 eV threshold, 86.33% of the time, while achieving a 1331x speedup in
+    computation. To standardize benchmarking, we introduce the Open Catalyst
+    Dense dataset containing nearly 1,000 diverse surfaces and 85,658 unique
+    configurations.
+  accessed:
+    - year: 2023
+      month: 8
+      day: 7
+  author:
+    - family: Lan
+      given: Janice
+    - family: Palizhati
+      given: Aini
+    - family: Shuaibi
+      given: Muhammed
+    - family: Wood
+      given: Brandon M.
+    - family: Wander
+      given: Brook
+    - family: Das
+      given: Abhishek
+    - family: Uyttendaele
+      given: Matt
+    - family: Zitnick
+      given: C. Lawrence
+    - family: Ulissi
+      given: Zachary W.
+  citation-key: lan_adsorbml_2023
+  issued:
+    - year: 2023
+      month: 1
+      day: 4
+  number: arXiv:2211.16486
+  publisher: arXiv
+  source: arXiv.org
+  title: 'AdsorbML: Accelerating Adsorption Energy Calculations with Machine Learning'
+  title-short: AdsorbML
+  type: article
+  URL: http://arxiv.org/abs/2211.16486
+
 - id: li_critical_2023
   abstract: >-
     Recent advances in machine learning (ML) methods have led to substantial

site/src/routes/preprint/references.yaml

@@ -1,4 +1,3 @@
-import os
 from glob import glob
 
 import yaml
@@ -18,8 +17,9 @@ def test_model_dirs_have_metadata() -> None:
         "repo",
     )
     for model_dir in MODEL_DIRS:
-        md_file = f"{model_dir}metadata.yml"
-        assert os.path.isfile(md_file), f"Missing metadata file: {md_file}"
+        md_files = glob(f"{model_dir}metadata*.yml")
+        assert len(md_files) == 1, f"expected 1 metadata file, got {md_files=}"
+        md_file = md_files[0]
 
         # make sure all required keys are non-empty
         with open(md_file) as yml_file: