Merge pull request #297 from ThomasBouche/feature/replace_3_6_by_3_10

support for python 3.10 and deletion of support 3.6

Author: guillaume-vignal

.github/workflows/main.yml

@@ -8,7 +8,7 @@ jobs:
     strategy:
       max-parallel: 1
       matrix:
-        python-version: [3.6, 3.7, 3.8, 3.9]
+        python-version: ["3.7", "3.8", "3.9", "3.10"]
     steps:
     - uses: actions/checkout@v2
     - name: Set up Python ${{ matrix.python-version }}
@@ -28,12 +28,12 @@ jobs:
         flake8 . --count --exit-zero --max-complexity=10 --max-line-length=127 --statistics
     - name: Test with pytest
       run: |
-        pytest
+        python -m pytest
     - name: Build python package
       run: |
         python setup.py sdist bdist_wheel
     - name: Deploy to PyPI
-      if: success() && startsWith(github.ref, 'refs/tags') && matrix.python-version == 3.9
+      if: success() && startsWith(github.ref, 'refs/tags') && matrix.python-version == '3.10'
       uses: pypa/gh-action-pypi-publish@master
       with:
         user: __token__

README.md

@@ -161,7 +161,7 @@ Shapash can use category-encoders object, sklearn ColumnTransformer or simply fe
 
 ## 🛠 Installation
 
-Shapash is intended to work with Python versions 3.6 to 3.9. Installation can be done with pip:
+Shapash is intended to work with Python versions 3.7 to 3.10. Installation can be done with pip:
 
 ```
 pip install shapash

requirements.dev.txt

@@ -1,18 +1,18 @@
 pip==21.3.1
+numpy>=1.18.0
 dash==2.3.1
-catboost==0.26.1
-category-encoders==2.2.2
+catboost>=1.0.1
+category-encoders>=2.2.2
 dash-bootstrap-components==1.1.0
 dash-core-components==2.0.0
 dash-daq==0.5.0
 dash-html-components==2.0.0
 dash-renderer==1.8.3
 dash-table==5.0.0
 lightgbm==2.3.1
-numpy>1.18.0
 pandas>1.0.2
 plotly==5.6.0
-shap==0.38.1
+shap>=0.38.1
 Sphinx==4.5.0
 sphinxcontrib-applehelp==1.0.2
 sphinxcontrib-devhelp==1.0.2
@@ -22,19 +22,20 @@ sphinxcontrib-qthelp==1.0.3
 sphinxcontrib-serializinghtml==1.1.5
 nbsphinx==0.8.8
 sphinx_material==0.0.35
-pytest==5.2.3
-pytest-cov==2.8.1
-scikit-learn==0.24.2
+pytest>=6.2.5
+pytest-cov>=2.8.1
+scikit-learn>=0.24.2
 xgboost==1.0.0
 nbformat>4.2.0
-numba==0.53.1
+numba>=0.53.1
 nbconvert>=6.0.7
-papermill
-matplotlib
+papermill>=2.0.0
+matplotlib>=3.3.0
 seaborn==0.11.1
-notebook
-Jinja2>=2.11.0,<3.1.0
-phik
-acv-exp==1.1.2
-lime
-regex
+notebook>=6.0.0
+Jinja2>=2.11.0
+phik>=0.12.0
+skranger>=0.6.0
+acv-exp>=1.1.2
+lime>=0.2.0.0
+regex

setup.py

@@ -41,7 +41,7 @@
 # This list should be identical to the list in shapash/report/__init__.py
 extras['report'] = [
     'nbconvert==6.0.7',
-    'papermill',
+    'papermill>=2.0.0',
     'seaborn<=0.11.2',
     'notebook',
     'Jinja2>=2.11.0,<3.1.0',
@@ -50,11 +50,11 @@
 
 extras['xgboost'] = ['xgboost>=1.0.0']
 extras['lightgbm'] = ['lightgbm>=2.3.0']
-extras['catboost'] = ['catboost>=0.21']
+extras['catboost'] = ['catboost>=1.0.1']
 extras['scikit-learn'] = ['scikit-learn>=0.23.0']
 extras['category_encoders'] = ['category_encoders>=2.2.2']
-extras['acv'] = ['acv-exp==1.1.2']
-extras['lime'] = ['lime']
+extras['acv'] = ['acv-exp==1.2.0']
+extras['lime'] = ['lime>=0.2.0.0']
 
 setup_requirements = ['pytest-runner', ]
 
@@ -63,19 +63,19 @@
 setup(
     name="shapash",
     version=version_d['__version__'],
-    python_requires='>3.5, <3.10',
+    python_requires='>3.6, <3.11',
     url='https://github.com/MAIF/shapash',
     author="Yann Golhen, Sebastien Bidault, Yann Lagre, Maxime Gendre",
     author_email="[email protected]",
     description="Shapash is a Python library which aims to make machine learning interpretable and understandable by everyone.",
     long_description=long_description,
     long_description_content_type='text/markdown',
     classifiers=[
-        "Programming Language :: Python :: 3",
-        'Programming Language :: Python :: 3.6',
+        "Programming Language :: Python :: 3",        
         'Programming Language :: Python :: 3.7',
         'Programming Language :: Python :: 3.8',
         'Programming Language :: Python :: 3.9',
+        'Programming Language :: Python :: 3.10',
         "License :: OSI Approved :: Apache Software License",
         "Operating System :: OS Independent",
     ],

shapash/utils/columntransformer_backend.py

@@ -465,7 +465,10 @@ def get_col_mapping_ct(encoder, x_encoded):
                 raise NotImplementedError(f'Estimator not supported : {estimator}')
 
         elif estimator == 'passthrough':
-            features_out = encoder._feature_names_in[features]
+            try:
+                features_out = encoder.feature_names_in_[features]
+            except:
+                features_out = encoder._feature_names_in[features] #for oldest sklearn version
             for f_name in features_out:
                 dict_col_mapping[f_name] = [x_encoded.columns.to_list()[idx_encoded]]
                 idx_encoded += 1

tests/data/predictor_to_load_310.pkl

@@ -661,15 +661,15 @@ def test_load_1(self):
         """
         temp, xpl = init_sme_to_pickle_test()
 
-        current = Path(path.abspath(__file__)).parent.parent.parent
-        if str(sys.version)[0:3] == '3.6':
-            pkl_file = path.join(current, 'data/xpl_to_load_36.pkl')
-        elif str(sys.version)[0:3] == '3.7':
+        current = Path(path.abspath(__file__)).parent.parent.parent        
+        if str(sys.version)[0:3] == '3.7':
             pkl_file = path.join(current, 'data/xpl_to_load_37.pkl')
         elif str(sys.version)[0:3] == '3.8':
             pkl_file = path.join(current, 'data/xpl_to_load_38.pkl')
         elif str(sys.version)[0:3] == '3.9':
             pkl_file = path.join(current, 'data/xpl_to_load_39.pkl')
+        elif str(sys.version)[0:4] == '3.10':
+            pkl_file = path.join(current, 'data/xpl_to_load_310.pkl')
         else:
             raise NotImplementedError
 
@@ -1152,7 +1152,7 @@ def test_generate_report(self, mock_execute_report, mock_export_and_save_report)
         mock_export_and_save_report.assert_called_once()
 
     def test_compute_features_stability_1(self):
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         selection = [1, 3]
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
@@ -1168,7 +1168,7 @@ def test_compute_features_stability_1(self):
         assert xpl.features_stability["amplitude"].shape == expected
 
     def test_compute_features_stability_2(self):
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         selection = [1]
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]

tests/unit_tests/explainer/test_smart_explainer.py

@@ -2064,7 +2064,7 @@ def test_correlations_2(self):
 
     def test_stability_plot_1(self):
         np.random.seed(42)
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
         model = DecisionTreeRegressor().fit(X, y)
@@ -2081,7 +2081,7 @@ def test_stability_plot_1(self):
 
     def test_stability_plot_2(self):
         np.random.seed(42)
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
         model = DecisionTreeRegressor().fit(X, y)
@@ -2101,7 +2101,7 @@ def test_stability_plot_2(self):
             assert np.array(list(output.data[0].x)).dtype == "float"
 
     def test_stability_plot_3(self):
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
         model = DecisionTreeRegressor().fit(X, y)
@@ -2120,7 +2120,7 @@ def test_stability_plot_3(self):
             assert np.array(list(output.data[0].x)).dtype == "float"
 
     def test_stability_plot_4(self):
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
         model = DecisionTreeRegressor().fit(X, y)
@@ -2140,7 +2140,7 @@ def test_stability_plot_4(self):
             assert np.array(list(output.data[0].x)).dtype == "float"
 
     def test_stability_plot_5(self):
-        df = pd.DataFrame(np.random.randint(0, 100, size=(15, 4)), columns=list('ABCD'))
+        df = pd.DataFrame(np.random.randint(1, 100, size=(15, 4)), columns=list('ABCD'))
         X = df.iloc[:, :-1]
         y = df.iloc[:, -1]
         model = DecisionTreeRegressor().fit(X, y)

tests/unit_tests/explainer/test_smart_plotter.py

@@ -8,7 +8,7 @@
 from sklearn.compose import ColumnTransformer
 import sklearn.preprocessing as skp
 import catboost as cb
-import sklearn
+import sklearn as sk
 import lightgbm
 import xgboost
 from shapash.utils.transform import inverse_transform
@@ -563,12 +563,19 @@ def test_inv_transform_ct_18(self):
         enc.fit(train, y)
         test = pd.DataFrame({'num1': [0, 1, 1],
                              'num2': [0, 2, 3],
-                             'other': ['A', 'B', 'C']})
-
-        expected = pd.DataFrame({'std_num1': [0.0, 1.0, 1.0],
-                                 'std_num2': [0.0, 2.0, 3.0],
-                                 'other': ['A', 'B', 'C']},
-                                dtype=object)
+                             'other': ['A', 'B', 'C']},
+                              )
+        if sk.__version__ >="1.0.0":
+            expected = pd.DataFrame({'std_num1': [0.0, 1.0, 1.0],
+                                    'std_num2': [0.0, 2.0, 3.0],
+                                    'other': ['A', 'B', 'C']},
+                                    )
+        else:
+            expected = pd.DataFrame({'std_num1': [0.0, 1.0, 1.0],
+                                    'std_num2': [0.0, 2.0, 3.0],
+                                    'other': ['A', 'B', 'C']},
+                                    dtype=object
+                                    )
 
         result = pd.DataFrame(enc.transform(test))
         result.columns = ['col1', 'col2', 'other']

tests/unit_tests/utils/test_columntransformer_backend.py

@@ -26,15 +26,16 @@ def test_load_smartpredictor_1(self):
         current = Path(path.abspath(__file__)).parent.parent.parent
         if str(sys.version)[0:3] == '3.7':
             pkl_file = path.join(current, 'data/predictor_to_load_37.pkl')
-        elif str(sys.version)[0:3] == '3.6':
-            pkl_file = path.join(current, 'data/predictor_to_load_36.pkl')
+        elif str(sys.version)[0:4] == '3.10':
+            pkl_file = path.join(current, 'data/predictor_to_load_310.pkl')
         elif str(sys.version)[0:3] == '3.8':
             pkl_file = path.join(current, 'data/predictor_to_load_38.pkl')
         elif str(sys.version)[0:3] == '3.9':
             pkl_file = path.join(current, 'data/predictor_to_load_39.pkl')
         else:
             raise NotImplementedError
 
+        predictor.save(pkl_file)
         predictor2 = load_smartpredictor(pkl_file)
 
         attrib_predictor = [element for element in predictor.__dict__.keys()]