Bug incorrect plot displayed after sorting table rows in webapp

.github/workflows/main.yml

@@ -8,7 +8,7 @@ jobs:
     strategy:
       max-parallel: 1
       matrix:
-        python-version: ["3.9", "3.10", "3.11", "3.12", "3.13"]
+        python-version: ["3.11", "3.12", "3.13", "3.14"]
     steps:
     - uses: actions/checkout@v3
     - name: Set up Python ${{ matrix.python-version }}

.pre-commit-config.yaml

@@ -33,10 +33,10 @@ repos:
         args: [--fix]
       - id: ruff-format
 -   repo: https://github.com/asottile/pyupgrade
-    rev: v2.7.2
+    rev: v3.21.2
     hooks:
     -   id: pyupgrade
-        args: [--py37-plus]
+        args: [--py311-plus]
 -   repo: https://github.com/asottile/blacken-docs
     rev: v1.8.0
     hooks:

README.md

@@ -142,7 +142,7 @@ Shapash can use category-encoders object, sklearn ColumnTransformer or simply fe
 
 ## 🛠 Installation
 
-Shapash is intended to work with Python versions 3.9 to 3.12. Installation can be done with pip:
+Shapash is intended to work with Python versions 3.11 to 3.14. Installation can be done with pip:
 
 ```bash
 pip install shapash

pyproject.toml

@@ -17,16 +17,15 @@ authors = [
 ]
 description = "Shapash is a Python library which aims to make machine learning interpretable and understandable by everyone."
 readme = "README.md"
-requires-python = ">=3.9, <3.14"
+requires-python = ">=3.11, <3.15"
 license = {text = "Apache Software License 2.0"}
 keywords = ["shapash"]
 classifiers = [
     "Programming Language :: Python :: 3",
-    "Programming Language :: Python :: 3.9",
-    "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
     "Programming Language :: Python :: 3.12",
     "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
     "License :: OSI Approved :: Apache Software License",
     "Operating System :: OS Independent",
 ]

shapash/backend/base_backend.py

@@ -1,5 +1,5 @@
 from abc import ABC, abstractmethod
-from typing import Any, Optional, Union
+from typing import Any
 
 import numpy as np
 import pandas as pd
@@ -28,7 +28,7 @@ class BaseBackend(ABC):
     support_groups = True
     supported_cases = ["classification", "regression"]
 
-    def __init__(self, model: Any, preprocessing: Optional[Any] = None):
+    def __init__(self, model: Any, preprocessing: Any | None = None):
         """Create a backend instance using a given implementation.
 
         Parameters
@@ -63,13 +63,12 @@ def run_explainer(self, x: pd.DataFrame) -> dict:
             dict containing local contributions
         """
         raise NotImplementedError(
-            f"`{self.__class__.__name__}` is a subclass of BaseBackend and "
-            f"must implement the `_run_explainer` method"
+            f"`{self.__class__.__name__}` is a subclass of BaseBackend and must implement the `_run_explainer` method"
         )
 
     def get_local_contributions(
-        self, x: pd.DataFrame, explain_data: Any, subset: Optional[list[int]] = None
-    ) -> Union[pd.DataFrame, list[pd.DataFrame]]:
+        self, x: pd.DataFrame, explain_data: Any, subset: list[int] | None = None
+    ) -> pd.DataFrame | list[pd.DataFrame]:
         """Get local contributions using the explainer data computed in the `run_explainer`
         method.
 
@@ -109,10 +108,10 @@ def get_local_contributions(
     def get_global_features_importance(
         self,
         contributions: pd.DataFrame,
-        explain_data: Optional[dict] = None,
-        subset: Optional[list[int]] = None,
+        explain_data: dict | None = None,
+        subset: list[int] | None = None,
         norm: int = 1,
-    ) -> Union[pd.Series, list[pd.Series]]:
+    ) -> pd.Series | list[pd.Series]:
         """Get global contributions using the explainer data computed in the `run_explainer`
         method.
 
@@ -141,8 +140,8 @@ def get_global_features_importance(
     def format_and_aggregate_local_contributions(
         self,
         x: pd.DataFrame,
-        contributions: Union[pd.DataFrame, np.array, list[pd.DataFrame], list[np.array]],
-    ) -> Union[pd.DataFrame, list[pd.DataFrame]]:
+        contributions: pd.DataFrame | np.ndarray | list[pd.DataFrame] | list[np.ndarray],
+    ) -> pd.DataFrame | list[pd.DataFrame]:
         """
         This function allows to format and aggregate contributions in the right format
         (pd.DataFrame or list of pd.DataFrame).
@@ -175,8 +174,8 @@ def format_and_aggregate_local_contributions(
         return contributions
 
     def _apply_preprocessing(
-        self, contributions: Union[pd.DataFrame, list[pd.DataFrame]]
-    ) -> Union[pd.DataFrame, list[pd.DataFrame]]:
+        self, contributions: pd.DataFrame | list[pd.DataFrame]
+    ) -> pd.DataFrame | list[pd.DataFrame]:
         """
         Reconstruct contributions for original features, taken into account a preprocessing.
 

shapash/decomposition/contributions.py

@@ -117,6 +117,6 @@ def assign_contributions(ranked):
     """
     if len(ranked) != 3:
         raise ValueError(
-            f"Expected lenght : 3, observed lenght : {len(ranked)}," "please check the outputs of rank_contributions."
+            f"Expected length : 3, observed length : {len(ranked)}, please check the outputs of rank_contributions."
         )
     return {"contrib_sorted": ranked[0], "x_sorted": ranked[1], "var_dict": ranked[2]}

shapash/explainer/consistency.py

@@ -30,7 +30,9 @@ def tuning_colorscale(self, values):
         desc_df = values.describe(percentiles=np.arange(0.1, 1, 0.1).tolist())
         min_pred, max_init = list(desc_df.loc[["min", "max"]].values)
         desc_pct_df = (desc_df.loc[~desc_df.index.isin(["count", "mean", "std"])] - min_pred) / (max_init - min_pred)
-        color_scale = list(map(list, (zip(desc_pct_df.values.flatten(), self._style_dict["init_contrib_colorscale"]))))
+        color_scale = list(
+            map(list, (zip(desc_pct_df.values.flatten(), self._style_dict["init_contrib_colorscale"], strict=False)))
+        )
         return color_scale
 
     def compile(self, contributions, x=None, preprocessing=None):
@@ -416,7 +418,9 @@ def plot_examples(self, method_1, method_2, l2, index, backend_name_1, backend_n
             axes = np.array([axes])
         fig.suptitle("Examples of explanations' comparisons for various distances (L2 norm)")
 
-        for n, (i, j, k, l, m, o) in enumerate(zip(method_1, method_2, l2, index, backend_name_1, backend_name_2)):
+        for n, (i, j, k, l, m, o) in enumerate(
+            zip(method_1, method_2, l2, index, backend_name_1, backend_name_2, strict=False)
+        ):
             # Only keep top features according to both methods
             idx = np.flip(np.abs(np.concatenate([i, j])).argsort()) % len(i)
             _, first_occurrence_idx = np.unique(idx, return_index=True)
@@ -601,6 +605,7 @@ def plot_pairwise_consistency(
                     weights[0][c].round(2),
                     weights[1][c].round(2),
                     (weights[0][c] - weights[1][c]).round(2),
+                    strict=False,
                 )
             ]
 

shapash/explainer/multi_decorator.py

@@ -76,7 +76,7 @@ def check_args(self, args, name):
         """
         if not args:
             raise ValueError(
-                f"{name} is applied without arguments," "please check that you have specified contributions."
+                f"{name} is applied without arguments, please check that you have specified contributions."
             )
 
     def check_method(self, method, name):
@@ -177,7 +177,7 @@ def combine_masks(self, masks):
         pd.Dataframe
             Combination of all masks.
         """
-        transposed_masks = list(map(list, zip(*masks)))
+        transposed_masks = list(map(list, zip(*masks, strict=False)))
         return self.delegate("combine_masks", transposed_masks)
 
     def compute_masked_contributions(self, s_contrib, masks):
@@ -197,7 +197,7 @@ def compute_masked_contributions(self, s_contrib, masks):
         list
             List of masked contributions (pandas.Series).
         """
-        arg_tup = list(zip(s_contrib, masks))
+        arg_tup = list(zip(s_contrib, masks, strict=False))
         return self.delegate("compute_masked_contributions", arg_tup)
 
     def summarize(self, s_contribs, var_dicts, xs_sorted, masks, columns_dict, features_dict):
@@ -224,7 +224,7 @@ def summarize(self, s_contribs, var_dicts, xs_sorted, masks, columns_dict, featu
         list of pd.DataFrame
             Result of the summarize step
         """
-        arg_tup = list(zip(s_contribs, var_dicts, xs_sorted, masks))
+        arg_tup = list(zip(s_contribs, var_dicts, xs_sorted, masks, strict=False))
         return self.delegate("summarize", arg_tup, columns_dict, features_dict)
 
     def compute_features_import(self, contributions, norm=1):

shapash/explainer/smart_plotter.py

@@ -4,7 +4,6 @@
 
 import math
 import random
-from typing import Optional
 
 import numpy as np
 import pandas as pd
@@ -452,7 +451,7 @@ def contribution_plot(
         if self._explainer._case == "classification":
             label_num, _, label_value = self._explainer.check_label_name(label)
 
-        if not isinstance(col, (str, int)):
+        if not isinstance(col, str | int):
             raise ValueError("parameter col must be string or int.")
         if hasattr(self._explainer, "inv_features_dict"):
             col = self._explainer.inv_features_dict.get(col, col)
@@ -977,7 +976,10 @@ def compare_plot(
                     f"Response: <b>{label_value}</b> - "
                     + "Probas: "
                     + " ; ".join(
-                        [str(id) + ": <b>" + str(round(proba, 2)) + "</b>" for proba, id in zip(preds, line_reference)]
+                        [
+                            str(id) + ": <b>" + str(round(proba, 2)) + "</b>"
+                            for proba, id in zip(preds, line_reference, strict=False)
+                        ]
                     )
                 )
 
@@ -988,7 +990,10 @@ def compare_plot(
             if show_predict:
                 preds = [self._explainer._local_pred(line) for line in line_reference]
                 subtitle = "Predictions: " + " ; ".join(
-                    [str(id) + ": <b>" + str(round(pred, 2)) + "</b>" for id, pred in zip(line_reference, preds)]
+                    [
+                        str(id) + ": <b>" + str(round(pred, 2)) + "</b>"
+                        for id, pred in zip(line_reference, preds, strict=False)
+                    ]
                 )
 
         new_contrib = list()
@@ -1006,11 +1011,11 @@ def compare_plot(
         preds = [self._explainer.x_init.loc[id] for id in line_reference]
         dict_features = self._explainer.inv_features_dict
 
-        iteration_list = list(zip(new_contrib, feature_values))
+        iteration_list = list(zip(new_contrib, feature_values, strict=False))
         iteration_list.sort(key=lambda x: maximum_difference_sort_value(x), reverse=True)
         iteration_list = iteration_list[:max_features]
         iteration_list = iteration_list[::-1]
-        new_contrib, feature_values = list(zip(*iteration_list))
+        new_contrib, feature_values = list(zip(*iteration_list, strict=False))
 
         fig = plot_line_comparison(
             line_reference,
@@ -1122,7 +1127,7 @@ def interactions_plot(
         >>> xpl.plot.interactions_plot(0, 1)
         """
 
-        if not (isinstance(col1, (str, int)) or isinstance(col2, (str, int))):
+        if not (isinstance(col1, str | int) or isinstance(col2, str | int)):
             raise ValueError("parameters col1 and col2 must be string or int.")
 
         col_id1 = self._explainer.check_features_name([col1])[0]
@@ -1925,7 +1930,7 @@ def confusion_matrix_plot(
     def distribution_plot(
         self,
         col: str,
-        hue: Optional[str] = None,
+        hue: str | None = None,
         width: int = 700,
         height: int = 500,
         nb_cat_max: int = 7,
@@ -2106,7 +2111,7 @@ def clustering_by_explainability_plot(
 
         if not hasattr(self._explainer, "model"):
             raise AssertionError(
-                "Explainer object was not compiled. Please compile the explainer " "object using .compile(...) method."
+                "Explainer object was not compiled. Please compile the explainer object using .compile(...) method."
             )
 
         if not hasattr(self._explainer, "_case") or self._explainer._case not in {"classification", "regression"}:

shapash/explainer/smart_state.py

@@ -43,7 +43,7 @@ def validate_contributions(self, contributions, x_init):
         pandas.DataFrame
             Local contributions on the original feature space (no encoding).
         """
-        if not isinstance(contributions, (np.ndarray, pd.DataFrame)):
+        if not isinstance(contributions, np.ndarray | pd.DataFrame):
             raise ValueError("Type of contributions must be pd.DataFrame or np.ndarray")
         if isinstance(contributions, np.ndarray):
             return pd.DataFrame(contributions, columns=x_init.columns, index=x_init.index)

shapash/manipulation/select_lines.py

@@ -49,10 +49,10 @@ def keep_right_contributions(y_pred, contributions, _case, _classes, label_dict,
 
     """
     if _case == "classification":
-        complete_sum = [list(x) for x in list(zip(*[df.values.tolist() for df in contributions]))]
+        complete_sum = [list(x) for x in list(zip(*[df.values.tolist() for df in contributions], strict=False))]
         indexclas = [_classes.index(x) for x in list(flatten(y_pred.values))]
         summary = pd.DataFrame(
-            [summar[ind] for ind, summar in zip(indexclas, complete_sum)],
+            [summar[ind] for ind, summar in zip(indexclas, complete_sum, strict=False)],
             columns=contributions[0].columns,
             index=contributions[0].index,
             dtype=object,
@@ -61,7 +61,7 @@ def keep_right_contributions(y_pred, contributions, _case, _classes, label_dict,
             y_pred = y_pred.map(lambda x: label_dict[x])
         if proba_values is not None:
             y_proba = pd.DataFrame(
-                [proba[ind] for ind, proba in zip(indexclas, proba_values.values)],
+                [proba[ind] for ind, proba in zip(indexclas, proba_values.values, strict=False)],
                 columns=["proba"],
                 index=y_pred.index,
             )

shapash/manipulation/summarize.py

@@ -99,7 +99,7 @@ def summarize(s_contrib, var_dict, x_sorted, mask, columns_dict, features_dict):
     summary = pd.concat([contrib_sum, var_dict_sum, x_sorted_sum], axis=1)
 
     # Ordering columns
-    ordered_columns = list(flatten(zip(var_dict_sum.columns, x_sorted_sum.columns, contrib_sum.columns)))
+    ordered_columns = list(flatten(zip(var_dict_sum.columns, x_sorted_sum.columns, contrib_sum.columns, strict=False)))
     summary = summary[ordered_columns]
     return summary
 

shapash/plots/plot_bar_chart.py

@@ -101,7 +101,7 @@ def plot_bar_chart(
             margin={"l": 150, "r": 20, "t": topmargin, "b": 70},
         )
         bars = []
-        for num, expl in enumerate(zip(var_dict, x_val, contrib)):
+        for num, expl in enumerate(zip(var_dict, x_val, contrib, strict=False)):
             feat_name, x_val_el, contrib_value = expl
             is_grouped = False
             if x_val_el == "":

shapash/plots/plot_contribution.py

@@ -89,7 +89,9 @@ def plot_scatter(
     feature_values = pd.DataFrame({column_name: feature_values_str})
 
     if pred is not None:
-        hv_text = [f"Id: {x}<br />Predict: {y}" for x, y in zip(feature_values.index, pred.values.flatten())]
+        hv_text = [
+            f"Id: {x}<br />Predict: {y}" for x, y in zip(feature_values.index, pred.values.flatten(), strict=False)
+        ]
     else:
         hv_text = [f"Id: {x}" for x in feature_values.index]
 
@@ -679,7 +681,9 @@ def _prepare_hover_text(feature_values, pred, feature_name):
     hv_text = [
         f"Id: {id_val}{f'<br />Predict: {pred_val}' if pred is not None else ''}"
         for id_val, pred_val in zip(
-            feature_values.index, pred.values.flatten() if pred is not None else [""] * len(feature_values)
+            feature_values.index,
+            pred.values.flatten() if pred is not None else [""] * len(feature_values),
+            strict=False,
         )
     ]
 

shapash/plots/plot_correlations.py

@@ -1,5 +1,3 @@
-from typing import Optional
-
 import numpy as np
 import pandas as pd
 import scipy.cluster.hierarchy as sch
@@ -14,7 +12,7 @@
 
 def plot_correlations(
     df,
-    style_dict: Optional[dict] = None,
+    style_dict: dict | None = None,
     palette_name: str = "default",
     features_dict=None,
     optimized=False,
@@ -197,7 +195,7 @@ def prepare_corr_matrix(df_subset):
                     coloraxis="coloraxis",
                     text=[
                         [
-                            f"Feature 1: {features_dict.get(y, y)} <br />" f"Feature 2: {features_dict.get(x, x)}"
+                            f"Feature 1: {features_dict.get(y, y)} <br />Feature 2: {features_dict.get(x, x)}"
                             for x in list_features
                         ]
                         for y in list_features
@@ -219,7 +217,7 @@ def prepare_corr_matrix(df_subset):
                 coloraxis="coloraxis",
                 text=[
                     [
-                        f"Feature 1: {features_dict.get(y, y)} <br />" f"Feature 2: {features_dict.get(x, x)}"
+                        f"Feature 1: {features_dict.get(y, y)} <br />Feature 2: {features_dict.get(x, x)}"
                         for x in list_features
                     ]
                     for y in list_features