Fix test scripts raising wandb.Table AssertionError: columns argu…

…ment expects list of strings or ints (#82)

* fix model test scripts raising wandb.Table AssertionError: columns argument expects list of strings or ints

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>

.gitignore

@@ -27,4 +27,3 @@ site/src/routes/api/*.md
 
 # temporary ignore rules
 2023-02-05-ens=10-perturb=5
-models/m3gnet/2023-05-26-*

.pre-commit-config.yaml

@@ -7,7 +7,7 @@ default_install_hook_types: [pre-commit, commit-msg]
 
 repos:
   - repo: https://github.com/astral-sh/ruff-pre-commit
-    rev: v0.1.13
+    rev: v0.1.15
     hooks:
       - id: ruff
         args: [--fix]
@@ -56,7 +56,7 @@ repos:
         exclude: ^(site/src/figs/.+\.svelte|data/wbm/20.+\..+|site/src/routes/.+\.(yaml|json)|changelog.md)$
 
   - repo: https://github.com/pre-commit/mirrors-eslint
-    rev: v9.0.0-alpha.0
+    rev: v9.0.0-alpha.2
     hooks:
       - id: eslint
         types: [file]

matbench_discovery/data.py

@@ -267,4 +267,5 @@ def _on_not_found(self, key: str, path: str) -> None:  # type: ignore[override]
 
 
 df_wbm = load("wbm_summary")
-df_wbm[Key.mat_id] = df_wbm.index
+# str() around Key.mat_id added for https://github.com/janosh/matbench-discovery/issues/81
+df_wbm[str(Key.mat_id)] = df_wbm.index

models/alignn_ff/test_alignn_ff.py

@@ -112,9 +112,7 @@
 
 
 # %%
-df_wbm = df_wbm.dropna()
-
-table = wandb.Table(dataframe=df_wbm[[Key.e_form, pred_col]].reset_index())
+table = wandb.Table(dataframe=df_wbm[[Key.e_form, pred_col]].reset_index().dropna())
 
 MAE = (df_wbm[Key.e_form] - df_wbm[pred_col]).abs().mean()
 R2 = r2_score(df_wbm[Key.e_form], df_wbm[pred_col])

models/chgnet/test_chgnet.py

@@ -136,7 +136,7 @@
 # %%
 df_wbm[e_pred_col] = df_out[e_pred_col]
 table = wandb.Table(
-    dataframe=df_wbm.dropna()[[Key.dft_energy, e_pred_col, Key.formula]].reset_index()
+    dataframe=df_wbm[[Key.dft_energy, e_pred_col, Key.formula]].reset_index().dropna()
 )
 
 title = f"CHGNet {task_type} ({len(df_out):,})"

models/mace/test_mace.py

@@ -38,10 +38,7 @@
 device = "cuda" if torch.cuda.is_available() else "cpu"
 # whether to record intermediate structures into pymatgen Trajectory
 record_traj = False  # has no effect if relax_cell is False
-model_name = [
-    "2023-10-29-mace-16M-pbenner-mptrj-no-conditional-loss",
-    "https://tinyurl.com/y7uhwpje",
-][-1]
+model_name = "https://tinyurl.com/y7uhwpje"
 ase_filter: Literal["frechet", "exp"] = "frechet"
 
 slurm_vars = slurm_submit(
@@ -163,8 +160,9 @@
 
 # %%
 df_wbm[e_pred_col] = df_out[e_pred_col]
+
 table = wandb.Table(
-    dataframe=df_wbm.dropna()[[Key.dft_energy, e_pred_col, Key.formula]].reset_index()
+    dataframe=df_wbm[[Key.dft_energy, e_pred_col, Key.formula]].reset_index().dropna()
 )
 
 title = f"MACE {task_type} ({len(df_out):,})"

scripts/model_figs/make_metrics_tables.py

@@ -175,7 +175,7 @@
     styler = (
         df_filtered.style.format(
             # render integers without decimal places
-            {key: "{:,.0f}" for key in "TP FN FP TN".split()},
+            dict.fromkeys("TP FN FP TN".split(), "{:,.0f}"),
             precision=2,  # render floats with 2 decimals
             na_rep="",  # render NaNs as empty string
         )