Merge pull request #64 from MAIF/feature/interaction_plot

Add interaction plot in the report

.github/workflows/main.yml

@@ -8,7 +8,7 @@ jobs:
     strategy:
       max-parallel: 1
       matrix:
-        python-version: ["3.9", "3.10", "3.11"]
+        python-version: ["3.9", "3.10", "3.11", "3.12"]
     steps:
     - uses: actions/checkout@v2
     - name: Set up Python ${{ matrix.python-version }}

README.md

@@ -107,7 +107,7 @@ sd.generate_report(
 
 ## 🛠 Installation
 
-Eurybia is intended to work with Python versions 3.9 to 3.11. Installation can be done with pip:
+Eurybia is intended to work with Python versions 3.9 to 3.12. Installation can be done with pip:
 
 ```
 pip install eurybia

eurybia/core/smartdrift.py

@@ -10,7 +10,6 @@
 import shutil
 import tempfile
 from pathlib import Path
-from typing import Dict
 
 import catboost
 import pandas as pd
@@ -199,12 +198,12 @@ def __init__(
     def compile(
         self,
         full_validation=False,
-        ignore_cols: list = list(),
+        ignore_cols: list = None,
         sampling=True,
         sample_size=100000,
         datadrift_file=None,
         date_compile_auc=None,
-        hyperparameter: Dict = catboost_hyperparameter_init.copy(),
+        hyperparameter: dict = catboost_hyperparameter_init.copy(),
         attr_importance="feature_importances_",
     ):
         r"""
@@ -237,6 +236,8 @@ def compile(
         >>> SD.compile()
 
         """
+        if ignore_cols is None:
+            ignore_cols = []
         if datadrift_file is not None:
             self.datadrift_file = datadrift_file
         if hyperparameter is not None:
@@ -468,7 +469,7 @@ def _analyze_consistency(self, full_validation=False, ignore_cols: list = list()
              and will not be analyzed: \n {err_dtypes}"""
             )
         # Feature values
-        err_mods: Dict[str, Dict] = {}
+        err_mods: dict[str, dict] = {}
         if full_validation is True:
             invalid_cols = ignore_cols + new_cols + removed_cols + err_dtypes
             for column in self.df_baseline.columns:

eurybia/core/smartplotter.py

@@ -597,7 +597,7 @@ def generate_modeldrift_data(
         if data_modeldrift is None:
             data_modeldrift = self.smartdrift.data_modeldrift
             if data_modeldrift is None:
-                raise Exception(
+                raise ValueError(
                     """You should run the add_data_modeldrift method before displaying model drift performances.
                 For more information see the documentation"""
                 )

eurybia/report/generation.py

@@ -232,7 +232,6 @@ def get_data_drift_panel(dr: DriftReport) -> pn.Column:
         pn.pane.Markdown("### Univariate analysis"),
         pn.pane.Markdown(report_text["Data drift"]["07"]),
     ]
-    contribution_figures, contribution_labels = dr.display_model_contribution()
 
     distribution_figures, labels, distribution_tables = dr.display_dataset_analysis(global_analysis=False)["univariate"]
     distribution_plots_blocks = get_select_plots(
@@ -262,6 +261,9 @@ def get_data_drift_panel(dr: DriftReport) -> pn.Column:
             max_gauge=0.2,
         )
         blocks += [pn.pane.Plotly(js_fig)]
+
+    contribution_figures, contribution_labels = dr.display_model_contribution()
+
     blocks += [
         pn.pane.Markdown("## Feature contribution on data drift's detection"),
         pn.pane.Markdown(report_text["Data drift"]["09"]),
@@ -273,14 +275,24 @@ def get_data_drift_panel(dr: DriftReport) -> pn.Column:
         figures=contribution_figures,
     )
     blocks += contribution_plots_blocks
+
+    fig_02 = dr.explainer.plot.top_interactions_plot(nb_top_interactions=10)
+    fig_02.update_layout(width=1240)
+    blocks += [
+        pn.pane.Markdown("## Feature interaction on data drift's detection"),
+        pn.pane.Markdown(report_text["Data drift"]["10"]),
+        pn.pane.Plotly(fig_02),
+    ]
+
     if dr.smartdrift.historical_auc is not None:
         fig = dr.smartdrift.plot.generate_historical_datadrift_metric()
         fig.update_layout(width=1240)
         blocks += [
             pn.pane.Markdown("## Historical Data drift"),
-            pn.pane.Markdown(report_text["Data drift"]["10"]),
+            pn.pane.Markdown(report_text["Data drift"]["11"]),
             pn.pane.Plotly(fig),
         ]
+
     return pn.Column(*blocks, name="Data drift", styles=dict(display="none"), css_classes=["data-drift"])
 
 

eurybia/report/project_report.py

@@ -5,7 +5,7 @@
 import copy
 import logging
 import os
-from typing import Dict, Optional, Union
+from typing import Optional, Union
 
 import jinja2
 import pandas as pd
@@ -36,11 +36,11 @@ class DriftReport:
     Attributes
     ----------
     smartdrift: object
-            SmartDrift object
+        SmartDrift object
     explainer : shapash.explainer.smart_explainer.SmartExplainer
-         A shapash SmartExplainer object that has already be compiled
+        A shapash SmartExplainer object that has already be compiled
     title_story : str
-            Report title
+        Report title
     metadata : dict
         Information about the project (author, description, ...)
     df_predict : pd.DataFrame
@@ -56,7 +56,7 @@ def __init__(
         smartdrift: SmartDrift,
         explainer: SmartExplainer,
         project_info_file: Optional[str] = None,
-        config_report: Optional[Dict] = None,
+        config_report: Optional[dict] = None,
     ):
         """
         Parameters
@@ -253,7 +253,7 @@ def display_model_contribution(self):
         c_list = self.explainer._classes if multiclass else [1]  # list just used for multiclass
         plot_list = []
         labels = []
-        for index_label, label in enumerate(c_list):  # Iterating over all labels in multiclass case
+        for label in c_list:  # Iterating over all labels in multiclass case
             for feature in self.features_imp_list:
                 fig = self.explainer.plot.contribution_plot(feature, label=label, max_points=200)
                 plot_list.append(fig)

eurybia/report/properties.py

@@ -1,6 +1,6 @@
-from typing import Any, Dict
+from typing import Any
 
-report_text: Dict[str, Any] = {
+report_text: dict[str, Any] = {
     "Index": {
         "01": "- Project information: report context and information",
         "02": "- Consistency Analysis: highlighting differences between the two datasets",
@@ -77,7 +77,12 @@
             "This representation constitutes a support to understand the drift "
             "when the analysis of the dataset is unclear."
         ),
-        "10": ("Line chart showing the metrics evolution of the datadrift classifier over the given period of time."),
+        "10": (
+            "This graph represents the interactions between couple of variable to the data drift detection. "
+            "This representation constitutes a support to understand the drift "
+            "when the analysis of the dataset is unclear."
+        ),
+        "11": ("Line chart showing the metrics evolution of the datadrift classifier over the given period of time."),
     },
     "Model drift": {
         "01": (

pyproject.toml

@@ -19,6 +19,7 @@ classifiers = [
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
     "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
     "License :: OSI Approved :: Apache Software License",
     "Operating System :: OS Independent",
 ]