spglib to moyo (#275)

* remove `get_cbar_label_formatter()` from `utils/plotting.py`

- Remove corresponding import and test cases

* import `calculate_rdf` in RDF module's `__init__.py`

- Update test case in `test_both_powerups.py` to use `add_scatter` instead of `add_trace`
- Minor SVG asset updates for element pair RDFs

* swap spglib for moyopy in example notebooks

plus minor refactors

* more spglib for moyopy swapping

- Update structure visualization scripts to use moyopy for space group number retrieval
- Modify spacegroup_sunburst and spacegroup_bar to work with moyopy
- Add mapping for Hermann-Maguin short symbols to space group numbers
- Remove commented-out SVG saving in sunburst script

* fix pymatgen SpaceGroup circular import error

materialsproject/pymatgen#4309

examples/matbench_perovskites_eda.ipynb remove ipynb cell output with huge line count

* breaking: rename `crystal_sys_from_spg_num()` to `spg_to_crystal_sys()`

- now also supports converting space group symbols to crystal systems, not just numbers
- Create `spg_num_to_from_symbol()` for bidirectional space group number/symbol conversion
- bump moyopy to 0.4.1
- Modify scripts to use new utility functions across various dataset exploration examples

.github/workflows/test.yml

@@ -24,7 +24,7 @@ jobs:
     with:
       os: ${{ matrix.os }}
       python-version: "3.10"
-      install-cmd: uv pip install -e '.[test,brillouin,symmetry]'  --system
+      install-cmd: uv pip install -e '.[test,brillouin]'  --system
       # pytest-split distributes work load so parallel jobs finish in similar time
       # update durations file with `pytest --store-durations --durations-path tests/files/.pytest-split-durations`
       test-cmd: pytest --durations 20 --cov-branch --cov-report=xml --cov pymatviz --splits 4 --group ${{ matrix.split }} --splitting-algorithm least_duration
@@ -61,7 +61,7 @@ jobs:
           python-version: "3.10"
 
       - name: Install package and dependencies
-        run: pip install -e .[make-assets,symmetry]
+        run: pip install -e .[make-assets]
 
       - name: Run script
         run: python ${{ matrix.script }}

assets/scripts/bar/spacegroup_bar.py

@@ -1,7 +1,6 @@
 # %%
+import pandas as pd
 from matminer.datasets import load_dataset
-from moyopy import SpaceGroupType
-from tqdm import tqdm
 
 import pymatviz as pmv
 from pymatviz.enums import Key
@@ -13,13 +12,12 @@
 # %%
 df_phonons = load_dataset("matbench_phonons")
 
-df_phonons[Key.spg_num] = [
-    struct.get_symmetry_dataset(backend="moyopy", return_raw_dataset=True).number
-    for struct in tqdm(df_phonons[Key.structure], desc="Getting spacegroups")
-]
-df_phonons[Key.spg_symbol] = [
-    SpaceGroupType(spg_num).hm_short for spg_num in df_phonons[Key.spg_num]
-]
+df_sym = pd.DataFrame(
+    struct.get_symmetry_dataset(backend="moyopy", return_raw_dataset=True).as_dict()
+    for struct in df_phonons[Key.structure]
+)
+df_phonons[Key.spg_num] = df_sym["number"]
+df_phonons[Key.spg_symbol] = df_sym["number"].map(pmv.utils.spg_num_to_from_symbol)
 
 
 # %% Spacegroup histograms

assets/scripts/ptable_plotly/ptable_heatmap_plotly.py

@@ -6,7 +6,6 @@
 from matminer.datasets import load_dataset
 
 import pymatviz as pmv
-from pymatviz import df_ptable
 from pymatviz.enums import ElemCountMode, Key
 
 
@@ -18,9 +17,9 @@
 
 # %% Plotly interactive periodic table heatmap
 fig = pmv.ptable_heatmap_plotly(
-    df_ptable[Key.atomic_mass],
+    pmv.df_ptable[Key.atomic_mass],
     hover_props=[Key.atomic_mass, Key.atomic_number],
-    hover_data="density = " + df_ptable[Key.density].astype(str) + " g/cm^3",
+    hover_data="density = " + pmv.df_ptable[Key.density].astype(str) + " g/cm^3",
     show_values=False,
 )
 fig.layout.title = dict(text="<b>Atomic mass heatmap</b>", x=0.4, y=0.94, font_size=20)
@@ -49,11 +48,11 @@
 
 # %% ex 1: Electronegativity Heatmap with Custom Hover Data
 fig = pmv.ptable_heatmap_plotly(
-    df_ptable[Key.electronegativity],
+    pmv.df_ptable[Key.electronegativity],
     colorscale="Viridis",
     hover_props=["atomic_mass", "melting_point"],
     hover_data={
-        el: f"Fun fact about {el}!" for el in df_ptable[Key.electronegativity].index
+        el: f"Fun fact about {el}!" for el in pmv.df_ptable[Key.electronegativity].index
     },
     font_colors=["white", "black"],
     colorbar=dict(title="Electronegativity", orientation="v"),
@@ -64,7 +63,7 @@
 
 # %% ex 2: Log-scale Abundance with Excluded Elements
 fig = pmv.ptable_heatmap_plotly(
-    df_ptable[Key.specific_heat_capacity].dropna(),
+    pmv.df_ptable[Key.specific_heat_capacity].dropna(),
     # colorscale="YlOrRd",
     log=True,
     font_colors=["black"],
@@ -80,7 +79,7 @@
 
 # %% ex 3: Fictional Data with Percent Mode and Custom Color Scale
 rand_data = {
-    elem: np.random.default_rng(seed=0).random() * 100 for elem in df_ptable.index
+    elem: np.random.default_rng(seed=0).random() * 100 for elem in pmv.df_ptable.index
 }
 custom_colorscale = [
     (0, "rgb(0,0,255)"),
@@ -116,10 +115,11 @@
 
 # %% ex 5: Atomic Radius with Custom Hover and Label Mapping
 fig = pmv.ptable_heatmap_plotly(
-    df_ptable[Key.atomic_radius],
+    pmv.df_ptable[Key.atomic_radius],
     colorscale="RdYlBu",
     hover_data={
-        elem: f"{radius} pm" for elem, radius in df_ptable[Key.atomic_radius].items()
+        elem: f"{radius} pm"
+        for elem, radius in pmv.df_ptable[Key.atomic_radius].items()
     },
     font_colors=["black"],
     colorbar=dict(title="Atomic Radius (pm)"),

assets/scripts/structure_viz/structure_2d.py

@@ -57,10 +57,10 @@
             site.species = {"Zr": 0.5, "Hf": 0.5}
 
     ax = cast(plt.Axes, pmv.structure_2d(struct))
-    _, spacegroup = struct.get_space_group_info()
+    spg_num = struct.get_symmetry_dataset(backend="moyopy")["number"]
 
     formula = struct.formula.replace(" ", "")
-    text = f"{formula}\ndisordered {mp_id}, {spacegroup = }"
+    text = f"{formula}\ndisordered {mp_id}, {spg_num = }"
     href = f"https://materialsproject.org/materials/{mp_id}"
     ax.text(
         0.5, 1, text, url=href, ha="center", transform=ax.transAxes, fontweight="bold"

assets/scripts/sunburst/spacegroup_sunburst.py

@@ -9,18 +9,25 @@
 # %%
 df_phonons = load_dataset("matbench_phonons")
 
-df_phonons[[Key.spg_symbol, Key.spg_num]] = [
-    struct.get_space_group_info() for struct in tqdm(df_phonons[Key.structure])
+df_phonons[Key.spg_num] = [
+    struct.get_symmetry_dataset(backend="moyopy", return_raw_dataset=True).number
+    for struct in tqdm(df_phonons[Key.structure])
 ]
 
+df_phonons[Key.spg_symbol] = df_phonons[Key.spg_num].map(
+    pmv.utils.spg_num_to_from_symbol
+)
+
 
 # %% Sunburst Plots
 fig = pmv.spacegroup_sunburst(df_phonons[Key.spg_num], show_counts="percent")
 title = "Matbench Phonons Spacegroup Sunburst"
 fig.layout.title = dict(text=f"<b>{title}</b>", x=0.5, y=0.96, font_size=18)
-pmv.io.save_and_compress_svg(fig, "spg-num-sunburst")
+fig.show()
+# pmv.io.save_and_compress_svg(fig, "spg-num-sunburst")
 
 fig = pmv.spacegroup_sunburst(df_phonons[Key.spg_symbol], show_counts="percent")
 title = "Matbench Phonons Spacegroup Symbols Sunburst"
 fig.layout.title = dict(text=f"<b>{title}</b>", x=0.5, y=0.96, font_size=18)
-pmv.io.save_and_compress_svg(fig, "spg-symbol-sunburst")
+fig.show()
+# pmv.io.save_and_compress_svg(fig, "spg-symbol-sunburst")

examples/dataset_exploration/camd_2022/explore_camd_2022.py

@@ -18,10 +18,8 @@
 
 import pandas as pd
 import requests
-from moyopy import SpaceGroupType
 
 import pymatviz as pmv
-from pymatviz.enums import Key
 
 
 # %% Download data (if needed)
@@ -56,8 +54,6 @@
 
 
 # %%
-df_camd[Key.spg_num] = [SpaceGroupType(spg).number for spg in df_camd.space_group]
-
-fig = pmv.spacegroup_sunburst(df_camd[Key.spg_num], show_counts="percent")
+fig = pmv.spacegroup_sunburst(df_camd.space_group, show_counts="percent")
 pmv.save_fig(fig, "camd-2022-spacegroup-sunburst.pdf")
 fig.show()

examples/dataset_exploration/matbench/dielectric/explore_dielectric.py

@@ -41,9 +41,7 @@
 ]
 df_diel[n_wyckoff_pos_key] = df_diel[proto_label_key].map(count_wyckoff_positions)
 
-df_diel[Key.crystal_system] = df_diel[Key.spg_num].map(
-    pmv.utils.crystal_sys_from_spg_num
-)
+df_diel[Key.crystal_system] = df_diel[Key.spg_num].map(pmv.utils.spg_to_crystal_sys)
 
 df_diel[Key.volume] = [x.volume for x in df_diel[Key.structure]]
 df_diel[Key.formula] = [x.formula for x in df_diel[Key.structure]]
@@ -61,7 +59,7 @@
 fig = pmv.spacegroup_bar(df_diel[Key.spg_num])
 fig.layout.title.update(text="<b>Space group histogram</b>")
 fig.layout.margin.update(b=10, l=10, r=10, t=50)
-# pmv.save_fig(ax, "dielectric-spacegroup-hist.pdf")
+# pmv.save_fig(fig, "dielectric-spacegroup-hist.pdf")
 fig.show()
 
 

examples/dataset_exploration/matbench/jdft2d/explore_jdft2d.py

@@ -35,7 +35,7 @@
 fig.layout.title.update(text=title, x=0.5)
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "jdft2d-exfoliation-energy-hist.pdf")
+# pmv.save_fig(fig, "jdft2d-exfoliation-energy-hist.pdf")
 
 
 # %%
@@ -56,12 +56,12 @@
 fig.layout.title.update(text=title, y=0.98)
 fig.layout.margin.update(b=10, l=10, r=10, t=70)
 fig.show()
-# pmv.save_fig(ax, "jdft2d-spacegroup-hist.pdf")
+# pmv.save_fig(fig, "jdft2d-spacegroup-hist.pdf")
 
 
 # %%
 fig = pmv.spacegroup_sunburst(df_2d[Key.spg_num], show_counts="percent")
 fig.layout.title.update(text="Spacegroup sunburst of the JARVIS DFT 2D dataset", x=0.5)
 fig.layout.margin.update(b=0, l=0, r=0, t=40)
 fig.show()
-# fig.write_image("jdft2d-spacegroup-sunburst.pdf")
+# pmv.save_fig(fig, "jdft2d-spacegroup-sunburst.pdf")

examples/dataset_exploration/matbench/log_g+kvrh/explore_log_g+krvh.py

@@ -13,10 +13,11 @@
 from time import perf_counter
 
 import numpy as np
+import pandas as pd
 import plotly.express as px
-from aviary.wren.utils import (
+from matbench_discovery.structure.prototype import (
     count_wyckoff_positions,
-    get_protostructure_label_from_spglib,
+    get_protostructure_label,
 )
 from matminer.datasets import load_dataset
 from pymatgen.core import Structure
@@ -30,36 +31,33 @@
 df_grvh = load_dataset("matbench_log_gvrh")
 df_kvrh = load_dataset("matbench_log_kvrh")
 
-# getting space group symbols and numbers for 10,987 structures takes about 45 sec
-df_grvh[[Key.spg_symbol, Key.spg_num]] = [
-    struct.get_space_group_info()
-    for struct in tqdm(
-        df_grvh[Key.structure], desc="Getting matbench_log_gvrh spacegroups"
-    )
-]
-df_grvh[Key.crystal_system] = df_grvh[Key.spg_num].map(
-    pmv.utils.crystal_sys_from_spg_num
+df_sym = pd.DataFrame(
+    struct.get_symmetry_dataset(backend="moyopy", return_raw_dataset=True).as_dict()
+    for struct in df_grvh[Key.structure]
 )
+df_sym[Key.crystal_system] = df_sym["number"].map(pmv.utils.spg_to_crystal_sys)
+df_grvh[Key.protostructure] = [
+    get_protostructure_label(struct)
+    for struct in tqdm(df_grvh[Key.structure], desc="matbench_log_gvrh Wyckoff strings")
+]
+df_kvrh[Key.protostructure] = df_grvh[Key.protostructure]
 
+for df in (df_grvh, df_kvrh):
+    df[[Key.spg_num, Key.wyckoff_symbols]] = df_sym[["number", "wyckoffs"]]
+    df[Key.crystal_system] = df_sym[Key.crystal_system]
 
-df_grvh[Key.wyckoff] = [
-    get_protostructure_label_from_spglib(struct)
-    for struct in tqdm(
-        df_grvh[Key.structure], desc="Getting matbench_log_gvrh Wyckoff strings"
-    )
-]
-df_grvh[Key.n_wyckoff_pos] = df_grvh.wyckoff.map(count_wyckoff_positions)
-df_grvh[Key.formula] = [x.formula for x in df_grvh[Key.structure]]
+    df[Key.n_wyckoff_pos] = df[Key.protostructure].map(count_wyckoff_positions)
+    df[Key.formula] = [x.formula for x in df[Key.structure]]
 
 
 # %%
 print("Number of materials with shear modulus of 0:")
-print(sum(df_grvh["log10(G_VRH)"] == 0))  # sum is 31
+print(sum(df_grvh["log10(G_VRH)"] == 0))  # expect 31
 
 
 # %%
 print("Number of materials with bulk modulus of 0:")
-print(sum(df_kvrh["log10(K_VRH)"] == 0))  # sum is 14
+print(sum(df_kvrh["log10(K_VRH)"] == 0))  # expect 14
 
 
 # %%
@@ -78,7 +76,7 @@
 fig.layout.title.update(text=title, x=0.5)
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "log_gvrh-volume-hist.pdf")
+# pmv.save_fig(fig, "gvrh-volume-hist.pdf")
 
 
 # %%
@@ -137,7 +135,7 @@ def has_isolated_atom(crystal: Structure, radius: float = 5) -> bool:
 title = "<b>Element counts in the Matbench<br>bulk/shear modulus datasets</b>"
 fig.layout.title.update(text=title)
 fig.show()
-# pmv.save_fig(fig, "log_gvrh-ptable-heatmap.pdf")
+# pmv.save_fig(fig, "gvrh-ptable-heatmap.pdf")
 
 
 # %%
@@ -147,14 +145,14 @@ def has_isolated_atom(crystal: Structure, radius: float = 5) -> bool:
     text="Spacegroup histogram of the Matbench bulk/shear modulus datasets"
 )
 fig.show()
-# pmv.save_fig(ax, "log_gvrh-spacegroup-hist.pdf")
+# pmv.save_fig(fig, "gvrh-spacegroup-hist.pdf")
 
 
 # %%
 fig = pmv.spacegroup_sunburst(df_grvh[Key.spg_num], show_counts="percent")
 title = "Spacegroup sunburst of the Matbench bulk/shear modulus datasets"
 fig.layout.title.update(text=title, x=0.5)
-# fig.write_image("log_gvrh-spacegroup-sunburst.pdf")
+# pmv.save_fig(fig, "gvrh-spacegroup-sunburst.pdf")
 fig.show()
 
 
@@ -199,4 +197,4 @@ def rgb_color(val: float, max_val: float) -> str:
 fig.layout.update(width=1000, height=400)
 fig.layout.xaxis = dict(tickvals=list(range(7)), ticktext=list(x_ticks.values()))
 fig.show()
-# fig.write_image("log_grvh-violin-num-wyckoffs.pdf")
+# pmv.save_fig(fig, "grvh-violin-num-wyckoffs.pdf")

examples/dataset_exploration/matbench/mp_e_form/explore_mp_e_form.py

@@ -28,7 +28,7 @@
 fig.layout.title.update(text=title, x=0.5)
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "mp_e_form_hist.pdf")
+# pmv.save_fig(fig, "mp-e-form-hist.pdf")
 
 
 # %%

examples/dataset_exploration/matbench/mp_gap/explore_mp_gap.py

@@ -31,7 +31,7 @@
 fig.layout.title.update(text=title, x=0.5)
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "pbe_gap_hist.pdf")
+# pmv.save_fig(fig, "pbe-gap-hist.pdf")
 
 
 # %%
@@ -47,4 +47,4 @@
 fig.layout.title.update(text=title, x=0.5)
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "volume_per_atom_hist.pdf")
+# pmv.save_fig(fig, "volume-per-atom-hist.pdf")

examples/dataset_exploration/matbench/perovskites/explore_perovskites.py

@@ -31,17 +31,15 @@
 
 df_perov[Key.formula] = df_perov[Key.structure].map(lambda cryst: cryst.formula)
 
-df_perov[Key.crystal_system] = df_perov[Key.spg_num].map(
-    pmv.utils.crystal_sys_from_spg_num
-)
+df_perov[Key.crystal_system] = df_perov[Key.spg_num].map(pmv.utils.spg_to_crystal_sys)
 
 
 # %%
 fig = pmv.structure_2d_plotly(df_perov[Key.structure].iloc[:12])
 fig.layout.title.update(text="Perovskite structures in Matbench Perovskites dataset")
 fig.layout.paper_bgcolor = "white"
 fig.show()
-# pmv.save_fig(ax, "perovskite-structures-2d.pdf")
+# pmv.save_fig(fig, "perovskite-structures-2d.pdf")
 
 
 # %%
@@ -51,7 +49,7 @@
 )
 fig.layout.showlegend = False
 fig.show()
-# pmv.save_fig(ax, "perovskites-e_form-hist.pdf")
+# pmv.save_fig(fig, "perovskites-e_form-hist.pdf")
 
 
 # %%
@@ -67,7 +65,7 @@
 pmv.powerups.annotate_bars(fig, v_offset=250)
 
 fig.show()
-# pmv.save_fig(ax, "perovskites-crystal-system-counts.pdf")
+# pmv.save_fig(fig, "perovskites-crystal-system-counts.pdf")
 
 
 # %%

examples/dataset_exploration/matbench/phonons/explore_phonons.py

@@ -38,7 +38,7 @@
     text="Last phonon DOS peak histogram of Matbench Phonons dataset"
 )
 fig.show()
-# pmv.save_fig(ax, "phonons-last-dos-peak-hist.pdf")
+# pmv.save_fig(fig, "phonons-last-dos-peak-hist.pdf")
 
 
 # %%