bella-companion 0.0.47__tar.gz → 0.0.48__tar.gz

{bella_companion-0.0.47 → bella_companion-0.0.48}/PKG-INFO

@@ -1,11 +1,11 @@
 Metadata-Version: 2.4
 Name: bella-companion
-Version: 0.0.47
+Version: 0.0.48
 Summary: Bayesian Evolutionary Layered Learning Architectures (BELLA) companion
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 Requires-Dist: arviz>=0.22.0
-Requires-Dist: bella-lumiere>=0.0.14
+Requires-Dist: bella-lumiere>=0.0.19
 Requires-Dist: dotenv>=0.9.9
 Requires-Dist: phylogenie>=3.1.4
 Requires-Dist: poetry>=2.2.1

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/eucovid/beast_configs/GLM.xml

@@ -143,7 +143,7 @@
         <operator spec="BactrianScaleOperator" parameter="@gammaShape" weight="1.0"/>
         <operator spec="BactrianScaleOperator" parameter="@kappa" weight="1.0"/>
 
-        <logger spec="Logger" fileName="MCMC.log" logEvery="1000" model="@posterior" sanitiseHeaders="true" sort="smart">
+        <logger spec="Logger" fileName="MCMC.log" logEvery="1000" model="@posterior">
             <log idref="posterior"/>
             <log idref="likelihood"/>
             <log idref="prior"/>

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/eucovid/beast_configs/MLP.xml

@@ -144,7 +144,7 @@
         <operator spec="BactrianScaleOperator" parameter="@gammaShape" weight="1.0"/>
         <operator spec="BactrianScaleOperator" parameter="@kappa" weight="1.0"/>
 
-        <logger spec="Logger" fileName="MCMC.log" logEvery="1000" model="@posterior" sanitiseHeaders="true" sort="smart">
+        <logger spec="Logger" fileName="MCMC.log" logEvery="1000" model="@posterior">
             <log idref="posterior"/>
             <log idref="likelihood"/>
             <log idref="prior"/>

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/eucovid/run_beast.py

@@ -1,5 +1,4 @@
 import os
-from glob import glob
 from pathlib import Path
 
 from bella_companion.utils import submit_job
@@ -17,19 +16,32 @@ def run_beast():
     msa_file = data_dir / "msa.fasta"
 
     predictors_dir = data_dir / "predictors"
-    all_predictor_files = ",".join(glob(str(predictors_dir / "all" / "*.tsv")))
+    all_predictor_files = [
+        "dist_7e_ls",
+        "osi_x_7e_ls",
+        "osi_y_7e_ls",
+        "pop_y_7e_ls",
+        "closedborders_7e_nt",
+        "sharedborders_7e_nt",
+        "pop_x_7e_ls",
+        "flight_7e_ls",
+        "flight_pop_x_7e_ls",
+    ]
+    all_predictor_files = ",".join(
+        [str(predictors_dir / "all" / f"{file}.csv") for file in all_predictor_files]
+    )
     all_predictors_data = " ".join(
         [
             f"-D msa_file={msa_file}",
-            f"-D changeTimesFile={predictors_dir / 'changetimes_all_7e.tsv'}",
+            f"-D changeTimesFile={predictors_dir / 'changetimes_all_7e.csv'}",
             f"-D predictorFiles={all_predictor_files}",
         ]
     )
     flight_predictor_data = " ".join(
         [
             f"-D msa_file={msa_file}",
-            f"-D changeTimesFile={predictors_dir / 'changetimes_flights_4e.tsv'}",
-            f"-D predictorFiles={predictors_dir / 'flight_pop_x_4e_ls.tsv'}",
+            f"-D changeTimesFile={predictors_dir / 'changetimes_flights_4e.csv'}",
+            f"-D predictorFiles={predictors_dir / 'flight_pop_x_4e_ls.csv'}",
         ]
     )
 
@@ -52,7 +64,6 @@ def run_beast():
             mem_per_cpu=12000,
         )
         """
-
         os.makedirs(output_dir / "all-predictors-GLM", exist_ok=True)
         submit_job(
             " ".join(
@@ -99,6 +110,7 @@ def run_beast():
             cpus=128,
             mem_per_cpu=12000,
         )
+
         """
         os.makedirs(output_dir / "flights-MLP", exist_ok=True)
         submit_job(

bella_companion-0.0.48/bella_companion/features.py

@@ -0,0 +1,13 @@
+from dataclasses import dataclass
+
+
+@dataclass
+class Feature:
+    is_binary: bool
+    is_relevant: bool | None = None
+
+    @property
+    def color(self) -> str:
+        if self.is_relevant is None:
+            return "purple"
+        return "#E74C3C" if self.is_relevant else "gray"

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/figures/epi_multitype.py

@@ -8,16 +8,16 @@ import numpy as np
 import polars as pl
 from lumiere.backend import relu, sigmoid
 
-from bella_companion.simulations.figures.explain import (
-    plot_partial_dependencies,
-    plot_shap_features_importance,
-)
 from bella_companion.simulations.scenarios.epi_multitype import (
     MIGRATION_PREDICTOR,
     MIGRATION_RATE_UPPER,
     MIGRATION_RATES,
     SCENARIO,
 )
+from bella_companion.utils import (
+    plot_partial_dependencies,
+    plot_shap_features_importance,
+)
 
 
 def _plot_predictions(log_summary: pl.DataFrame, output_dir: Path):

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/figures/fbd_2traits.py

@@ -8,10 +8,6 @@ import numpy as np
 import polars as pl
 from lumiere.backend import relu, sigmoid
 
-from bella_companion.simulations.figures.explain import (
-    plot_partial_dependencies,
-    plot_shap_features_importance,
-)
 from bella_companion.simulations.figures.utils import step
 from bella_companion.simulations.scenarios.fbd_2traits import (
     FBD_RATE_UPPER,
@@ -20,6 +16,10 @@ from bella_companion.simulations.scenarios.fbd_2traits import (
     SCENARIO,
     STATES,
 )
+from bella_companion.utils import (
+    plot_partial_dependencies,
+    plot_shap_features_importance,
+)
 
 
 def _plot_predictions(log_summary: pl.DataFrame, output_dir: Path):

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/scenarios/epi_multitype.py

@@ -2,7 +2,7 @@ import numpy as np
 from numpy.random import Generator
 from phylogenie import get_epidemiological_events
 
-from bella_companion.simulations.features import Feature
+from bella_companion.features import Feature
 from bella_companion.simulations.scenarios.common import (
     BECOME_UNINFECTIOUS_RATE,
     EPI_MAX_TIME,

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/scenarios/epi_skyline.py

@@ -3,7 +3,7 @@ from functools import partial
 import numpy as np
 from phylogenie import SkylineParameter, get_epidemiological_events
 
-from bella_companion.simulations.features import Feature
+from bella_companion.features import Feature
 from bella_companion.simulations.scenarios.common import (
     BECOME_UNINFECTIOUS_RATE,
     EPI_MAX_TIME,

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/scenarios/fbd_2traits.py

@@ -2,7 +2,7 @@ import numpy as np
 from numpy.random import Generator
 from phylogenie import SkylineVector, get_canonical_events
 
-from bella_companion.simulations.features import Feature
+from bella_companion.features import Feature
 from bella_companion.simulations.scenarios.common import (
     FBD_MAX_TIME,
     FBD_RATE_UPPER,

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/scenarios/fbd_no_traits.py

@@ -3,7 +3,7 @@ from functools import partial
 import numpy as np
 from phylogenie import SkylineParameter, get_canonical_events
 
-from bella_companion.simulations.features import Feature
+from bella_companion.features import Feature
 from bella_companion.simulations.scenarios.common import (
     FBD_MAX_TIME,
     FBD_RATE_UPPER,

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/simulations/scenarios/scenario.py

@@ -6,7 +6,7 @@ from typing import Any
 from numpy.random import Generator
 from phylogenie.treesimulator import Event
 
-from bella_companion.simulations.features import Feature
+from bella_companion.features import Feature
 
 
 class ScenarioType(Enum):

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion/utils/__init__.py

@@ -4,8 +4,8 @@ from bella_companion.utils.beast import (
     summarize_logs_dir,
 )
 from bella_companion.utils.explain import (
-    get_median_partial_dependence_values,
-    get_median_shap_features_importance,
+    plot_partial_dependencies,
+    plot_shap_features_importance,
 )
 from bella_companion.utils.slurm import submit_job
 
@@ -13,7 +13,7 @@ __all__ = [
     "summarize_log",
     "summarize_logs_dir",
     "read_weights_dir",
-    "get_median_partial_dependence_values",
-    "get_median_shap_features_importance",
+    "plot_partial_dependencies",
+    "plot_shap_features_importance",
     "submit_job",
 ]

bella_companion-0.0.48/bella_companion/utils/explain/__init__.py

@@ -0,0 +1,7 @@
+from bella_companion.utils.explain.pdp import plot_partial_dependencies
+from bella_companion.utils.explain.shap import plot_shap_features_importance
+
+__all__ = [
+    "plot_partial_dependencies",
+    "plot_shap_features_importance",
+]

{bella_companion-0.0.47/bella_companion/simulations/figures → bella_companion-0.0.48/bella_companion/utils}/explain/pdp.py

@@ -6,20 +6,44 @@ import matplotlib.pyplot as plt
 import numpy as np
 import seaborn as sns
 from joblib import Parallel, delayed
-from lumiere.typing import ActivationFunction, Weights
+from lumiere import get_partial_dependence_values
+from lumiere.typing import ActivationFunction, ArrayLike, Weights
 from tqdm import tqdm
 
-from bella_companion.simulations.features import Feature
-from bella_companion.utils import get_median_partial_dependence_values
+import bella_companion.utils.explain.typeguards as tg
+from bella_companion.features import Feature
+
+
+def _get_median_partial_dependence_values(
+    weights: list[Weights],  # shape: (n_weight_samples, ...)
+    inputs: ArrayLike,
+    features_grid: list[list[float]],
+    hidden_activation: ActivationFunction,
+    output_activation: ActivationFunction,
+) -> list[list[float]]:  # shape: (n_features, n_grid_points)
+    pdvalues = [
+        get_partial_dependence_values(
+            weights=w,
+            features_grid=features_grid,
+            hidden_activation=hidden_activation,
+            output_activation=output_activation,
+        )
+        for w in weights
+    ]
+    return [
+        np.median([pd[feature_idx] for pd in pdvalues], axis=0).tolist()
+        for feature_idx in range(len(features_grid))
+    ]
 
 
 def plot_partial_dependencies(
-    weights: list[list[Weights]],  # shape: (n_runs, n_weights_samples, ...)
+    weights: list[Weights] | list[list[Weights]],
     features: dict[str, Feature],
-    output_dir: Path,
+    output_dir: str | Path,
     hidden_activation: ActivationFunction,
     output_activation: ActivationFunction,
 ):
+    output_dir = Path(output_dir)
     os.makedirs(output_dir, exist_ok=True)
 
     continuous_grid: list[float] = np.linspace(0, 1, 10).tolist()
@@ -27,10 +51,20 @@ def plot_partial_dependencies(
         [0, 1] if feature.is_binary else continuous_grid
         for feature in features.values()
     ]
+
+    if tg.is_weights_list(weights):
+        func = get_partial_dependence_values
+    elif tg.is_nested_weights_list(weights):
+        func = _get_median_partial_dependence_values
+    else:
+        raise ValueError(
+            "weights must be a list of Weights or a list of list of Weights"
+        )
+
     jobs = Parallel(n_jobs=-1, return_as="generator_unordered")(
         delayed(
             partial(
-                get_median_partial_dependence_values,
+                func,
                 features_grid=features_grid,
                 hidden_activation=hidden_activation,
                 output_activation=output_activation,
@@ -48,19 +82,22 @@ def plot_partial_dependencies(
     if any(not f.is_binary for f in features.values()):
         for (label, feature), feature_pdvalues in zip(features.items(), pdvalues):
             if not feature.is_binary:
-                color = "#E74C3C" if feature.is_relevant else "gray"
                 median = np.median(feature_pdvalues, axis=1)
                 lower = np.percentile(feature_pdvalues, 2.5, axis=1)
                 high = np.percentile(feature_pdvalues, 100 - 2.5, axis=1)
                 plt.fill_between(  # pyright: ignore
-                    continuous_grid, lower, high, alpha=0.25, color=color
+                    continuous_grid, lower, high, alpha=0.25, color=feature.color
                 )
                 for mcmc_pds in feature_pdvalues.T:
                     plt.plot(  # pyright: ignore
-                        continuous_grid, mcmc_pds, color=color, alpha=0.2, linewidth=1
+                        continuous_grid,
+                        mcmc_pds,
+                        color=feature.color,
+                        alpha=0.2,
+                        linewidth=1,
                     )
                 plt.plot(  # pyright: ignore
-                    continuous_grid, median, color=color, label=label
+                    continuous_grid, median, color=feature.color, label=label
                 )
         plt.xlabel("Feature value")  # pyright: ignore
         plt.ylabel("MLP Output")  # pyright: ignore
@@ -84,10 +121,9 @@ def plot_partial_dependencies(
         )
         ax.get_legend().remove()  # pyright: ignore
 
-        for i, f in enumerate([f for f in features.values() if f.is_binary]):
-            color = "#E74C3C" if f.is_relevant else "gray"
+        for i, feature in enumerate([f for f in features.values() if f.is_binary]):
             for coll in ax.collections[i * 2 : i * 2 + 2]:
-                coll.set_facecolor(color)
+                coll.set_facecolor(feature.color)
 
         plt.xlabel("Feature")  # pyright: ignore
         plt.ylabel("MLP Output")  # pyright: ignore

bella_companion-0.0.48/bella_companion/utils/explain/shap.py

@@ -0,0 +1,102 @@
+from functools import partial
+from itertools import product
+from pathlib import Path
+from typing import Callable
+
+import matplotlib.pyplot as plt
+import numpy as np
+import seaborn as sns
+from joblib import Parallel, delayed
+from lumiere import get_shap_features_importance
+from lumiere.typing import ActivationFunction, Array, Weights
+from numpy.typing import ArrayLike
+from tqdm import tqdm
+
+import bella_companion.utils.explain.typeguards as tg
+from bella_companion.features import Feature
+
+
+def _get_median_shap_features_importance(
+    weights: list[Weights],
+    inputs: ArrayLike,
+    hidden_activation: ActivationFunction,
+    output_activation: ActivationFunction,
+) -> list[float]:  # length: n_features
+    features_importance = np.array(
+        [
+            get_shap_features_importance(
+                weights=w,
+                inputs=inputs,
+                hidden_activation=hidden_activation,
+                output_activation=output_activation,
+            )
+            for w in weights
+        ]
+    )
+    return np.median(features_importance, axis=0).tolist()
+
+
+def _plot_shap_violins(
+    features: dict[str, Feature],
+    features_importance: Array,  # shape: (n_samples, n_features)
+    output_file: str | Path,
+):
+    features_importance = np.asarray(features_importance)
+    features_importance /= features_importance.sum(axis=1, keepdims=True)
+
+    def _plot_violins(group_check: Callable[[Feature], bool]):
+        for i, (feature_name, feature) in enumerate(features.items()):
+            if group_check(feature):
+                sns.violinplot(
+                    y=features_importance[:, i],
+                    x=[feature_name] * len(features_importance),
+                    cut=0,
+                    color=feature.color,
+                )
+
+    _plot_violins(lambda f: not f.is_binary)
+    _plot_violins(lambda f: f.is_binary)
+    plt.xlabel("Feature")  # pyright: ignore
+    plt.ylabel("Importance")  # pyright: ignore
+    plt.savefig(output_file)  # pyright: ignore
+    plt.close()
+
+
+def plot_shap_features_importance(
+    weights: list[Weights] | list[list[Weights]],
+    features: dict[str, Feature],
+    output_file: str | Path,
+    hidden_activation: ActivationFunction,
+    output_activation: ActivationFunction,
+):
+    continuous_grid: list[float] = np.linspace(0, 1, 10).tolist()
+    features_grid: list[list[float]] = [
+        [0, 1] if feature.is_binary else continuous_grid
+        for feature in features.values()
+    ]
+    inputs = list(product(*features_grid))
+
+    if tg.is_weights_list(weights):
+        func = get_shap_features_importance
+    elif tg.is_nested_weights_list(weights):
+        func = _get_median_shap_features_importance
+    else:
+        raise ValueError(
+            "weights must be a list of Weights or a list of list of Weights"
+        )
+
+    jobs = Parallel(n_jobs=-1, return_as="generator_unordered")(
+        delayed(
+            partial(
+                func,
+                inputs=inputs,
+                hidden_activation=hidden_activation,
+                output_activation=output_activation,
+            )
+        )(w)
+        for w in weights
+    )
+    features_importance = np.array(
+        [job for job in tqdm(jobs, total=len(weights), desc="Evaluating SHAPs")]
+    )  # shape: (n_runs, n_features)
+    _plot_shap_violins(features, features_importance, output_file)

bella_companion-0.0.48/bella_companion/utils/explain/typeguards.py

@@ -0,0 +1,20 @@
+from typing import Any, TypeGuard
+
+import numpy as np
+from lumiere.typing import Weights
+
+
+def is_list(obj: Any) -> TypeGuard[list[Any]]:
+    return isinstance(obj, list)
+
+
+def is_weights(obj: Any) -> TypeGuard[Weights]:
+    return is_list(obj) and all(isinstance(layer, np.ndarray) for layer in obj)
+
+
+def is_weights_list(weights: Any) -> TypeGuard[list[Weights]]:
+    return is_list(weights) and all(is_weights(w) for w in weights)
+
+
+def is_nested_weights_list(weights: Any) -> TypeGuard[list[list[Weights]]]:
+    return is_list(weights) and all(is_weights_list(w) for w in weights)

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion.egg-info/PKG-INFO

@@ -1,11 +1,11 @@
 Metadata-Version: 2.4
 Name: bella-companion
-Version: 0.0.47
+Version: 0.0.48
 Summary: Bayesian Evolutionary Layered Learning Architectures (BELLA) companion
 Requires-Python: >=3.10
 Description-Content-Type: text/markdown
 Requires-Dist: arviz>=0.22.0
-Requires-Dist: bella-lumiere>=0.0.14
+Requires-Dist: bella-lumiere>=0.0.19
 Requires-Dist: dotenv>=0.9.9
 Requires-Dist: phylogenie>=3.1.4
 Requires-Dist: poetry>=2.2.1

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion.egg-info/SOURCES.txt

@@ -3,6 +3,7 @@ pyproject.toml
 bella_companion/BELLA.jar
 bella_companion/__init__.py
 bella_companion/cli.py
+bella_companion/features.py
 bella_companion/version.xml
 bella_companion.egg-info/PKG-INFO
 bella_companion.egg-info/SOURCES.txt
@@ -12,6 +13,7 @@ bella_companion.egg-info/requires.txt
 bella_companion.egg-info/top_level.txt
 bella_companion/__pycache__/__init__.cpython-310.pyc
 bella_companion/__pycache__/cli.cpython-310.pyc
+bella_companion/__pycache__/features.cpython-310.pyc
 bella_companion/eucovid/__init__.py
 bella_companion/eucovid/run_beast.py
 bella_companion/eucovid/__pycache__/__init__.cpython-310.pyc
@@ -19,22 +21,22 @@ bella_companion/eucovid/__pycache__/run_beast.cpython-310.pyc
 bella_companion/eucovid/beast_configs/GLM.xml
 bella_companion/eucovid/beast_configs/MLP.xml
 bella_companion/eucovid/beast_configs/Nonparametric.xml
-bella_companion/eucovid/data/case_data.tsv
-bella_companion/eucovid/data/change_dates.tsv
+bella_companion/eucovid/data/case_data.csv
+bella_companion/eucovid/data/change_dates.csv
 bella_companion/eucovid/data/msa.fasta
-bella_companion/eucovid/data/osi.tsv
-bella_companion/eucovid/data/predictors/changetimes_all_7e.tsv
-bella_companion/eucovid/data/predictors/changetimes_flights_4e.tsv
-bella_companion/eucovid/data/predictors/flight_pop_x_4e_ls.tsv
-bella_companion/eucovid/data/predictors/all/closedborders_7e_nt.tsv
-bella_companion/eucovid/data/predictors/all/dist_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/flight_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/flight_pop_x_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/osi_x_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/osi_y_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/pop_x_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/pop_y_7e_ls.tsv
-bella_companion/eucovid/data/predictors/all/sharedborders_7e_nt.tsv
+bella_companion/eucovid/data/osi.csv
+bella_companion/eucovid/data/predictors/changetimes_all_7e.csv
+bella_companion/eucovid/data/predictors/changetimes_flights_4e.csv
+bella_companion/eucovid/data/predictors/flight_pop_x_4e_ls.csv
+bella_companion/eucovid/data/predictors/all/closedborders_7e_nt.csv
+bella_companion/eucovid/data/predictors/all/dist_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/flight_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/flight_pop_x_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/osi_x_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/osi_y_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/pop_x_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/pop_y_7e_ls.csv
+bella_companion/eucovid/data/predictors/all/sharedborders_7e_nt.csv
 bella_companion/platyrrhine/__init__.py
 bella_companion/platyrrhine/beast_config.xml
 bella_companion/platyrrhine/results.py
@@ -49,7 +51,6 @@ bella_companion/platyrrhine/data/change_times.csv
 bella_companion/platyrrhine/data/traits.csv
 bella_companion/platyrrhine/data/trees.nwk
 bella_companion/simulations/__init__.py
-bella_companion/simulations/features.py
 bella_companion/simulations/generate_data.py
 bella_companion/simulations/generate_figures.py
 bella_companion/simulations/metrics.py
@@ -89,12 +90,6 @@ bella_companion/simulations/figures/__pycache__/fbd_2traits.cpython-310.pyc
 bella_companion/simulations/figures/__pycache__/fbd_no_traits.cpython-310.pyc
 bella_companion/simulations/figures/__pycache__/scenarios.cpython-310.pyc
 bella_companion/simulations/figures/__pycache__/utils.cpython-310.pyc
-bella_companion/simulations/figures/explain/__init__.py
-bella_companion/simulations/figures/explain/pdp.py
-bella_companion/simulations/figures/explain/shap.py
-bella_companion/simulations/figures/explain/__pycache__/__init__.cpython-310.pyc
-bella_companion/simulations/figures/explain/__pycache__/pdp.cpython-310.pyc
-bella_companion/simulations/figures/explain/__pycache__/shap.cpython-310.pyc
 bella_companion/simulations/scenarios/__init__.py
 bella_companion/simulations/scenarios/common.py
 bella_companion/simulations/scenarios/epi_multitype.py
@@ -111,9 +106,16 @@ bella_companion/simulations/scenarios/__pycache__/fbd_no_traits.cpython-310.pyc
 bella_companion/simulations/scenarios/__pycache__/scenario.cpython-310.pyc
 bella_companion/utils/__init__.py
 bella_companion/utils/beast.py
-bella_companion/utils/explain.py
 bella_companion/utils/slurm.py
 bella_companion/utils/__pycache__/__init__.cpython-310.pyc
 bella_companion/utils/__pycache__/beast.cpython-310.pyc
 bella_companion/utils/__pycache__/explain.cpython-310.pyc
-bella_companion/utils/__pycache__/slurm.cpython-310.pyc
+bella_companion/utils/__pycache__/slurm.cpython-310.pyc
+bella_companion/utils/explain/__init__.py
+bella_companion/utils/explain/pdp.py
+bella_companion/utils/explain/shap.py
+bella_companion/utils/explain/typeguards.py
+bella_companion/utils/explain/__pycache__/__init__.cpython-310.pyc
+bella_companion/utils/explain/__pycache__/pdp.cpython-310.pyc
+bella_companion/utils/explain/__pycache__/shap.cpython-310.pyc
+bella_companion/utils/explain/__pycache__/typeguards.cpython-310.pyc

{bella_companion-0.0.47 → bella_companion-0.0.48}/bella_companion.egg-info/requires.txt

@@ -1,5 +1,5 @@
 arviz>=0.22.0
-bella-lumiere>=0.0.14
+bella-lumiere>=0.0.19
 dotenv>=0.9.9
 phylogenie>=3.1.4
 poetry>=2.2.1

{bella_companion-0.0.47 → bella_companion-0.0.48}/pyproject.toml

@@ -1,12 +1,12 @@
 [project]
 name = "bella-companion"
-version = "0.0.47"
+version = "0.0.48"
 description = "Bayesian Evolutionary Layered Learning Architectures (BELLA) companion"
 readme = "README.md"
 requires-python = ">=3.10"
 dependencies = [
     "arviz>=0.22.0",
-    "bella-lumiere>=0.0.14",
+    "bella-lumiere>=0.0.19",
     "dotenv>=0.9.9",
     "phylogenie>=3.1.4",
     "poetry>=2.2.1",
@@ -32,5 +32,3 @@ build-backend = "setuptools.build_meta"
 [tool.setuptools]
 packages = ["bella_companion"]
 
-[tool.setuptools.package-data]
-bella_companion = ["**/*"]

bella_companion-0.0.47/bella_companion/simulations/features.py

@@ -1,7 +0,0 @@
-from dataclasses import dataclass
-
-
-@dataclass
-class Feature:
-    is_binary: bool
-    is_relevant: bool

bella_companion-0.0.47/bella_companion/simulations/figures/explain/__init__.py

@@ -1,6 +0,0 @@
-from bella_companion.simulations.figures.explain.pdp import plot_partial_dependencies
-from bella_companion.simulations.figures.explain.shap import (
-    plot_shap_features_importance,
-)
-
-__all__ = ["plot_partial_dependencies", "plot_shap_features_importance"]

bella_companion-0.0.47/bella_companion/simulations/figures/explain/shap.py

@@ -1,62 +0,0 @@
-from functools import partial
-from itertools import product
-from pathlib import Path
-from typing import Callable
-
-import matplotlib.pyplot as plt
-import numpy as np
-import seaborn as sns
-from joblib import Parallel, delayed
-from lumiere.typing import ActivationFunction, Weights
-from tqdm import tqdm
-
-from bella_companion.simulations.features import Feature
-from bella_companion.utils import get_median_shap_features_importance
-
-
-def plot_shap_features_importance(
-    weights: list[list[Weights]],  # shape: (n_runs, n_weights_samples, ...)
-    features: dict[str, Feature],
-    output_file: Path,
-    hidden_activation: ActivationFunction,
-    output_activation: ActivationFunction,
-):
-    continuous_grid: list[float] = np.linspace(0, 1, 10).tolist()
-    features_grid: list[list[float]] = [
-        [0, 1] if feature.is_binary else continuous_grid
-        for feature in features.values()
-    ]
-    inputs = list(product(*features_grid))
-
-    jobs = Parallel(n_jobs=-1, return_as="generator_unordered")(
-        delayed(
-            partial(
-                get_median_shap_features_importance,
-                inputs=inputs,
-                hidden_activation=hidden_activation,
-                output_activation=output_activation,
-            )
-        )(w)
-        for w in weights
-    )
-    features_importances = np.array(
-        [job for job in tqdm(jobs, total=len(weights), desc="Evaluating SHAPs")]
-    )  # shape: (n_runs, n_features)
-    features_importances /= features_importances.sum(axis=1, keepdims=True)
-
-    def _plot_violins(group_check: Callable[[Feature], bool]):
-        for i, (feature_name, feature) in enumerate(features.items()):
-            if group_check(feature):
-                sns.violinplot(
-                    y=features_importances[:, i],
-                    x=[feature_name] * len(features_importances),
-                    cut=0,
-                    color="#E74C3C" if feature.is_relevant else "gray",
-                )
-
-    _plot_violins(lambda f: not f.is_binary)
-    _plot_violins(lambda f: f.is_binary)
-    plt.xlabel("Feature")  # pyright: ignore
-    plt.ylabel("Importance")  # pyright: ignore
-    plt.savefig(output_file)  # pyright: ignore
-    plt.close()

bella_companion-0.0.47/bella_companion/utils/explain.py

@@ -1,45 +0,0 @@
-import numpy as np
-from lumiere import get_partial_dependence_values, get_shap_features_importance
-from lumiere.typing import ActivationFunction, Weights
-from numpy.typing import ArrayLike
-
-
-def get_median_partial_dependence_values(
-    weights: list[Weights],  # shape: (n_weight_samples, ...)
-    features_grid: list[list[float]],
-    hidden_activation: ActivationFunction,
-    output_activation: ActivationFunction,
-) -> list[list[float]]:  # shape: (n_features, n_grid_points)
-    pdvalues = [
-        get_partial_dependence_values(
-            weights=w,
-            features_grid=features_grid,
-            hidden_activation=hidden_activation,
-            output_activation=output_activation,
-        )
-        for w in weights
-    ]
-    return [
-        np.median([pd[feature_idx] for pd in pdvalues], axis=0).tolist()
-        for feature_idx in range(len(features_grid))
-    ]
-
-
-def get_median_shap_features_importance(
-    weights: list[Weights],
-    inputs: ArrayLike,
-    hidden_activation: ActivationFunction,
-    output_activation: ActivationFunction,
-) -> list[float]:  # length: n_features
-    features_importance = np.array(
-        [
-            get_shap_features_importance(
-                weights=w,
-                inputs=inputs,
-                hidden_activation=hidden_activation,
-                output_activation=output_activation,
-            )
-            for w in weights
-        ]
-    )
-    return np.median(features_importance, axis=0).tolist()