- updated pipeline

- updated versions and tests

Author: maxsch3

.github/workflows/pipeline.yml

@@ -0,0 +1,26 @@
+name: Python package
+
+on: [push]
+
+jobs:
+  test:
+
+    runs-on: ubuntu-latest
+    strategy:
+      matrix:
+        python-version: ["3.8", "3.9", "3.10", "3.11"]
+
+    steps:
+      - uses: actions/checkout@v4
+      - name: Set up Python
+        uses: actions/setup-python@v4
+        with:
+          python-version: '3.x'
+      - name: Install dependencies
+        run: |
+          python -m pip install --upgrade pip
+          pip install -r requirements.txt
+      - name: Test with pytest
+        run: |
+          pip install pytest pytest-cov pytest-dependency
+          pytest tests --doctest-modules --junitxml=junit/test-results.xml --cov=com --cov-report=xml --cov-report=html

.github/workflows/test-pipeline.yml

@@ -0,0 +1,3 @@
+pandas>=2.1.4
+scikit-learn
+tensorflow>=2.10.0

requirements.txt

@@ -17,10 +17,9 @@
       download_url='https://github.com/maxsch3/batchflow',
       license='MIT',
       setup_requires=['setuptools_scm'],
-      install_requires=['numpy>=1.9.1',
-                        'scipy>=0.14',
+      install_requires=['numpy>=1.20.0',
                         'scikit-learn',
-                        'pandas'],
+                        'pandas>=2.0.0'],
       extras_require={
           'visualize': ['pydot>=1.2.4'],
           'tests': ['pytest',
@@ -33,7 +32,7 @@
           'Intended Audience :: Science/Research',
           'License :: OSI Approved :: MIT License',
           'Programming Language :: Python :: 3',
-          'Programming Language :: Python :: 3.6',
+          'Programming Language :: Python :: 3.10',
           'Topic :: Software Development :: Libraries',
           'Topic :: Software Development :: Libraries :: Python Modules'
       ],

requirements_tf1.txt

@@ -41,10 +41,11 @@ def test_cols_dist(self):
         assert p > 0.001
 
     def test_uniform_col_dist(self):
+        sample_size = 1000
         fd = FeatureDropout([0., 1.], ['var1', 'var2', 'label'], drop_values='')
-        batch = fd.transform(self.df.sample(1000, replace=True))
+        batch = fd.transform(self.df.sample(sample_size, replace=True))
         b = (batch == '').sum(axis=0)
-        c, p = chisquare(b, [333, 333, 333])
+        c, p = chisquare(b, f_exp=[sample_size/3, sample_size/3, sample_size/3])
         assert p > 0.01
 
     def test_different_drop_values(self):

requirements_tf2.txt

@@ -20,14 +20,16 @@ def test_transform(self):
 
     def test_transform_integer_array(self):
         """
-        This tests that pandas specific IntegerArray is converted into numpy format
+        This tests that pandas specific IntegerArray is converted into numpy format.
+        IntergerArray of type "Int64" in pandas is a very handy integer data type which supports Nones and does
+        not require conversion to float data type
         :return:
         """
         data = pd.Series([1, 2, 4, 5], dtype="Int64")
         nea = NumpyEncoderAdaptor()
         tr = nea.transform(data)
         assert isinstance(tr, np.ndarray)
-        assert np.issubdtype(tr.dtype, np.object)
+        assert np.issubdtype(tr.dtype, object)
 
     def test_transform_datetime(self):
         """
@@ -45,7 +47,7 @@ def test_inverse_transform(self):
         nea = NumpyEncoderAdaptor()
         tr = nea.inverse_transform(data)
         assert isinstance(tr, pd.Series)
-        assert np.issubdtype(tr.dtype, np.int)
+        assert np.issubdtype(tr.dtype, np.int64)
         tr = nea.inverse_transform(data, dtype=np.float32)
         assert isinstance(tr, pd.Series)
         assert np.issubdtype(tr.dtype, np.float32)

setup.py

@@ -34,7 +34,7 @@ def test_inverse_transform(self):
         pea = PandasEncoderAdaptor()
         tr = pea.inverse_transform(data)
         assert isinstance(tr, pd.Series)
-        assert np.issubdtype(tr.dtype, np.int)
+        assert np.issubdtype(tr.dtype, np.int64)
         tr = pea.inverse_transform(data, dtype=np.float32)
         assert isinstance(tr, pd.Series)
         assert np.issubdtype(tr.dtype, np.float32)