for #185: fix FileNotFoundError from Model.<key>.label when pip insta…

…ll . (i.e. non-editable install)

- add 'build' directory to .gitignore
- fix pyproject.toml to include model YAML files in package data
- remove more matplotlib remnants with plotly in data/wbm/compare_cse_vs_ce_mp_2020_corrections.py

.gitignore

@@ -1,6 +1,7 @@
 # package install
 *.egg-info
 dist
+build
 
 # cache
 __pycache__
@@ -14,7 +15,6 @@ __pycache__
 data/**/raw
 data/**/tsne
 !data/mp/2023-02-07-mp-elemental-reference-entries.json.gz
-models/**/checkpoints
 data/**/*.extxyz*
 data/**/*.json*
 data/**/*.zip*
@@ -34,7 +34,6 @@ models/**/*.tgz
 site/src/routes/api/*.md
 
 # temporary ignore rules
-models/cgcnn/2023-02-05-ens=10-perturb=5
 data/mp/mptrj-gga-ggapu/*
 
 # large files

data/wbm/compare_cse_vs_ce_mp_2020_corrections.py

@@ -1,13 +1,12 @@
-"""NOTE MaterialsProject2020Compatibility takes structural information into account when
-correcting energies (for oxides and sulfides only). Always use
-ComputedStructureEntry, not ComputedEntry when applying energy corrections.
+"""MaterialsProject2020Compatibility takes structural information into account when
+correcting energies (for oxides and sulfides only). Always pass
+ComputedStructureEntry, not ComputedEntry when calling process_entries.
 """
 
 # %%
 import gzip
 import json
 
-import matplotlib.pyplot as plt
 import pandas as pd
 import pymatviz as pmv
 from pymatgen.entries.compatibility import (
@@ -28,11 +27,17 @@
 
 cses = [
     ComputedStructureEntry.from_dict(dct)
-    for dct in tqdm(df_cse[Key.computed_structure_entry])
+    for dct in tqdm(
+        df_cse[Key.computed_structure_entry],
+        desc="Loading ComputedStructureEntries",
+    )
 ]
 
 ces = [
-    ComputedEntry.from_dict(dct) for dct in tqdm(df_cse[Key.computed_structure_entry])
+    ComputedEntry.from_dict(dct)
+    for dct in tqdm(
+        df_cse[Key.computed_structure_entry], desc="Loading ComputedEntries"
+    )
 ]
 
 
@@ -43,10 +48,14 @@
 assert len(processed) == len(df_cse)
 
 df_wbm["e_form_per_atom_mp2020_from_ce"] = [
-    get_e_form_per_atom(entry) for entry in tqdm(ces)
+    get_e_form_per_atom(entry)
+    for entry in tqdm(ces, desc="Calculating formation energies from ComputedEntries")
 ]
 df_wbm["e_form_per_atom_mp2020_from_cse"] = [
-    get_e_form_per_atom(entry) for entry in tqdm(cses)
+    get_e_form_per_atom(entry)
+    for entry in tqdm(
+        cses, desc="Calculating formation energies from ComputedStructureEntries"
+    )
 ]
 
 df_wbm["mp2020_cse_correction_per_atom"] = [
@@ -79,50 +88,46 @@
 df_wbm["anion"][df_wbm[Key.chem_sys].astype(str).str.contains("'O'")] = "oxide"
 df_wbm["anion"][df_wbm[Key.chem_sys].astype(str).str.contains("'S'")] = "sulfide"
 
-assert dict(df_wbm.anion.value_counts()) == {"oxide": 26984, "sulfide": 10606}
+anion_counts = dict(df_wbm.anion.value_counts())
+assert anion_counts == {"oxide": 26_984, "sulfide": 10_596}, f"{anion_counts=}"
 
 df_ce_ne_cse = df_wbm.query(
     "abs(e_form_per_atom_mp2020_from_cse - e_form_per_atom_mp2020_from_ce) > 1e-4"
 )
 
 
 # %%
-ax = plt.gca()
-for key, df_anion in df_ce_ne_cse.groupby("anion"):
-    ax = df_anion.plot.scatter(
-        ax=ax,
-        x="mp2020_cse_correction_per_atom",
-        y="mp2020_ce_correction_per_atom",
-        label=f"{key} ({len(df_anion):,})",
-        color=dict(oxide="orange", sulfide="teal").get(key, "blue"),
-        title=f"CSE vs CE corrections for ({len(df_ce_ne_cse):,} / {len(df_wbm):,} = "
-        f"{len(df_ce_ne_cse) / len(df_wbm):.1%})\n outliers of largest difference",
+for x_col, y_col, title in (
+    ("mp2020_cse_correction_per_atom", "mp2020_ce_correction_per_atom", "correction"),
+    ("e_form_per_atom_mp2020_from_cse", "e_form_per_atom_mp2020_from_ce", "e-form"),
+):
+    fig = df_ce_ne_cse.plot.scatter(
+        x=x_col,
+        y=y_col,
+        color="anion",
+        color_discrete_map={"oxide": "orange", "sulfide": "teal"},
+        hover_data=[Key.formula],
+        backend="plotly",
     )
-
-ax.axline((0, 0), slope=1, color="gray", linestyle="dashed", zorder=-1)
-
-pmv.save_fig(ax, f"{ROOT}/tmp/{today}-ce-vs-cse-corrections-outliers.pdf")
-
-
-# %%
-ax = plt.gca()
-for key, df_anion in df_ce_ne_cse.groupby("anion"):
-    ax = df_anion.plot.scatter(
-        ax=ax,
-        x="e_form_per_atom_mp2020_from_cse",
-        y="e_form_per_atom_mp2020_from_ce",
-        label=f"{key} ({len(df_anion):,})",
-        color=dict(oxide="orange", sulfide="teal").get(key, "blue"),
-        title=f"Outliers in formation energy from CSE vs CE ({len(df_ce_ne_cse):,}"
-        f" / {len(df_wbm):,} = {len(df_ce_ne_cse) / len(df_wbm):.1%})",
-    )
-
-ax.axline((0, 0), slope=1, color="gray", linestyle="dashed", zorder=-1)
-
-# insight: all materials for which ComputedEntry and ComputedStructureEntry give
-# different formation energies are oxides or sulfides for which MP 2020 compat takes
-# into account structural information to make more accurate corrections.
-pmv.save_fig(ax, f"{ROOT}/tmp/{today}-ce-vs-cse-e-form-outliers.pdf")
+    title = f"CSE vs CE {title}<br>({len(df_ce_ne_cse):,} / {len(df_wbm):,} "
+    title += f"= {len(df_ce_ne_cse) / len(df_wbm):.1%})"
+    fig.layout.title.update(text=title, x=0.5, font=dict(size=16))
+    fig.layout.margin.t = 40
+    fig.layout.legend.update(x=0, title=None)
+
+    pmv.powerups.add_identity_line(fig)
+
+    # Update legend labels with counts
+    for trace in fig.data:
+        anion = trace.name
+        count = len(df_ce_ne_cse[df_ce_ne_cse.anion == anion])
+        trace.name = f"{anion} ({count:,})"
+
+    # insight: all materials for which ComputedEntry and ComputedStructureEntry give
+    # different formation energies are oxides or sulfides for which MP 2020 compat takes
+    # into account structural information to make more accurate corrections.
+    pmv.save_fig(fig, f"{ROOT}/tmp/{today}-ce-vs-cse-{title}-outliers.pdf")
+    fig.show()
 
 
 # %% below code resulted in

matbench_discovery/__init__.py

@@ -19,7 +19,7 @@
 
 PKG_DIR = os.path.dirname(__file__)
 # repo root directory if editable install, else the pkg directory
-ROOT = os.path.dirname(PKG_DIR) if pkg_is_editable else PKG_DIR
+ROOT = os.path.dirname(PKG_DIR)
 DATA_DIR = f"{ROOT}/data"  # directory to store raw data
 SITE_FIGS = f"{ROOT}/site/src/figs"  # directory for interactive figures
 # directory to write model analysis for website

matbench_discovery/data.py

@@ -222,11 +222,6 @@ def __new__(
         obj._base_dir = base_dir  # noqa: SLF001
         return obj
 
-    @property
-    def member_map(cls: type[T]) -> dict[str, "Files"]:  # type: ignore[misc]
-        """Map of member names to member objects."""
-        return cls._member_map_  # type: ignore[return-value]
-
     @property
     def base_dir(cls) -> str:
         """Base directory of the file."""
@@ -504,7 +499,8 @@ def phonons_path(self) -> str | None:
         return f"{ROOT}/{rel_path}"
 
 
-px.defaults.labels |= {k: v.label for k, v in Model.member_map.items()}
+# render model keys as labels in plotly axes and legends
+px.defaults.labels |= {k.name: k.label for k in Model}
 
 
 def load_df_wbm_with_preds(
@@ -545,7 +541,7 @@ def load_df_wbm_with_preds(
     valid_models = {model.name for model in Model}
     if models == ():
         models = tuple(valid_models)
-    inv_label_map = {v.label: k for k, v in Model.member_map.items()}
+    inv_label_map = {key.label: key.name for key in Model}
     # map pretty model names back to Model enum keys
     models = {inv_label_map.get(model, model) for model in models}
     if unknown_models := ", ".join(models - valid_models):

models/chgnet/join_chgnet_preds.py

@@ -64,7 +64,7 @@
 
 
 # %%
-ax = pmv.density_scatter_plotly(
+pmv.density_scatter_plotly(
     df=df_preds,
     x=MbdKey.e_form_dft,
     y=e_form_chgnet_col,

pyproject.toml

@@ -69,7 +69,7 @@ include = ["matbench_discovery*"]
 exclude = ["tests", "tests.*"]
 
 [tool.setuptools.package-data]
-matbench_discovery = ["**/*.yml"]
+matbench_discovery = ["**/*.yml", "../models/**/*.yml"]
 
 [build-system]
 requires = ["setuptools>=70"]