make release-tag: Merge branch 'master' into stable

Author: sarahmish

HISTORY.md

@@ -1,5 +1,16 @@
 # History
 
+## 0.3.1 - 2021-10-07
+
+### Primitive Improvements
+
+* Dynamic target_shape in keras adapter - [Issue #263](https://github.com/MLBazaar/MLPrimitives/issues/263) by @sarahmish
+* Save keras primitives in Windows environment - [Issue #261](https://github.com/MLBazaar/MLPrimitives/issues/261) by @sarahmish
+
+### General Imporvements
+
+* Update TensorFlow and NumPy dependency - [Issue #259](https://github.com/MLBazaar/MLPrimitives/issues/259) by @sarahmish
+
 ## 0.3.0 - 2021-01-09
 
 ### New Primitives

mlprimitives/__init__.py

@@ -4,7 +4,7 @@
 
 __author__ = 'MIT Data To AI Lab'
 __email__ = 'dailabmit@gmail.com'
-__version__ = '0.3.0'
+__version__ = '0.3.1.dev1'
 
 import os
 

mlprimitives/adapters/keras.py

@@ -28,14 +28,14 @@ class Sequential(object):
     def __getstate__(self):
         state = self.__dict__.copy()
 
-        with tempfile.NamedTemporaryFile(suffix='.hdf5', delete=True) as fd:
+        with tempfile.NamedTemporaryFile(suffix='.hdf5', delete=False) as fd:
             keras.models.save_model(state.pop('model'), fd.name, overwrite=True)
             state['model_str'] = fd.read()
 
         return state
 
     def __setstate__(self, state):
-        with tempfile.NamedTemporaryFile(suffix='.hdf5', delete=True) as fd:
+        with tempfile.NamedTemporaryFile(suffix='.hdf5', delete=False) as fd:
             fd.write(state.pop('model_str'))
             fd.flush()
 
@@ -85,17 +85,18 @@ def _setdefault(self, kwargs, key, value):
         if key in self.hyperparameters and self.hyperparameters[key] is None:
             kwargs[key] = value
 
-    def _augment_hyperparameters(self, X, kwargs):
+    def _augment_hyperparameters(self, X, mode, kwargs):
         shape = np.asarray(X)[0].shape
         length = shape[0]
-        self._setdefault(kwargs, 'input_shape', shape)
-        self._setdefault(kwargs, 'input_dim', length)
-        self._setdefault(kwargs, 'input_length', length)
+        self._setdefault(kwargs, '{}_shape'.format(mode), shape)
+        self._setdefault(kwargs, '{}_dim'.format(mode), length)
+        self._setdefault(kwargs, '{}_length'.format(mode), length)
 
         return kwargs
 
     def fit(self, X, y, **kwargs):
-        self._augment_hyperparameters(X, kwargs)
+        self._augment_hyperparameters(X, 'input', kwargs)
+        self._augment_hyperparameters(y, 'target', kwargs)
         self.model = self._build_model(**kwargs)
 
         if self.classification:

setup.cfg

@@ -1,5 +1,5 @@
 [bumpversion]
-current_version = 0.3.0
+current_version = 0.3.1.dev1
 commit = True
 tag = True
 parse = (?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\.(?P<release>[a-z]+)(?P<candidate>\d+))?

setup.py

@@ -12,23 +12,23 @@
     history = history_file.read()
 
 install_requires = [
-    'Keras>=2.4',
+    'Keras>=2.4,<2.5',
     'featuretools>=0.6.1,<0.23',
     'iso639>=0.1.4,<0.2',
     'langdetect>=1.0.7,<2',
     'lightfm>=1.15,<2',
     'mlblocks>=0.4.0.dev0,<0.5',
     'networkx>=2.0,<3',
     'nltk>=3.3,<4',
-    'numpy<1.19.0,>=1.16.0',
+    'numpy<1.21.0,>=1.16.0',
     'opencv-python>=3.4.0.12,<5',
     'pandas>=1,<2',
     'python-louvain>=0.10,<0.14',   # community
     'scikit-image>=0.15',
     'scikit-learn>=0.21',
     'scipy>=1.1.0,<2',
     'statsmodels>=0.9.0,<0.13',
-    'tensorflow>=2,<2.4',
+    'tensorflow>=2,<2.5',
     'xgboost>=0.72.1,<1',
 ]
 
@@ -117,6 +117,6 @@
     test_suite='tests',
     tests_require=tests_require,
     url='https://github.com/MLBazaar/MLPrimitives',
-    version='0.3.0',
+    version='0.3.1.dev1',
     zip_safe=False,
 )

tests/adapters/test_keras.py

@@ -65,7 +65,7 @@ def test__augment_hyperparameters_3d_numpy(self):
              [1, 2, 3, 4],
              [1, 2, 3, 4]],
         ])
-        Sequential._augment_hyperparameters(sequential, X, kwargs)
+        Sequential._augment_hyperparameters(sequential, X, 'input', kwargs)
 
         # Assert
         assert kwargs == {'input_shape': (3, 4)}

tests/adapters/test_pandas.py

@@ -17,7 +17,7 @@ def setUp(self):
                 for day in range(4)
                 for hour in range(24)
             ],
-            'value': list(range(4 * 24))
+            'value': np.arange(0, 4 * 24, dtype=float)
         })
 
     def test_resample_rule_str(self):
@@ -50,22 +50,22 @@ def test_resample_groupby(self):
         out = resample(self.df.set_index('dt'), '1d', groupby=['group1', 'group2'])
 
         assert_frame_equal(out, pd.DataFrame([
-            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 1), 'value': 10},
-            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 2), 'value': 34},
-            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 3), 'value': 58},
-            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 4), 'value': 82},
-            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 1), 'value': 12},
-            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 2), 'value': 36},
-            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 3), 'value': 60},
-            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 4), 'value': 84},
-            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 1), 'value': 11},
-            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 2), 'value': 35},
-            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 3), 'value': 59},
-            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 4), 'value': 83},
-            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 1), 'value': 13},
-            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 2), 'value': 37},
-            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 3), 'value': 61},
-            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 4), 'value': 85},
+            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 1), 'value': 10.0},
+            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 2), 'value': 34.0},
+            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 3), 'value': 58.0},
+            {'group1': 'A', 'group2': 'C', 'dt': datetime(2000, 1, 4), 'value': 82.0},
+            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 1), 'value': 12.0},
+            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 2), 'value': 36.0},
+            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 3), 'value': 60.0},
+            {'group1': 'A', 'group2': 'D', 'dt': datetime(2000, 1, 4), 'value': 84.0},
+            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 1), 'value': 11.0},
+            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 2), 'value': 35.0},
+            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 3), 'value': 59.0},
+            {'group1': 'B', 'group2': 'C', 'dt': datetime(2000, 1, 4), 'value': 83.0},
+            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 1), 'value': 13.0},
+            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 2), 'value': 37.0},
+            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 3), 'value': 61.0},
+            {'group1': 'B', 'group2': 'D', 'dt': datetime(2000, 1, 4), 'value': 85.0},
         ], columns=['group1', 'group2', 'dt', 'value']))
 
     def test_resample_on(self):
@@ -95,30 +95,30 @@ def test_resample_aggregation_str(self):
         out = resample(self.df.set_index('dt'), '1d', aggregation='max')
 
         assert_frame_equal(out, pd.DataFrame([
-            {'dt': datetime(2000, 1, 1), 'value': 23},
-            {'dt': datetime(2000, 1, 2), 'value': 47},
-            {'dt': datetime(2000, 1, 3), 'value': 71},
-            {'dt': datetime(2000, 1, 4), 'value': 95},
+            {'dt': datetime(2000, 1, 1), 'value': 23.0},
+            {'dt': datetime(2000, 1, 2), 'value': 47.0},
+            {'dt': datetime(2000, 1, 3), 'value': 71.0},
+            {'dt': datetime(2000, 1, 4), 'value': 95.0},
         ]))
 
     def test_resample_aggregation_func(self):
 
         out = resample(self.df.set_index('dt'), '1d', aggregation=np.max)
 
         assert_frame_equal(out, pd.DataFrame([
-            {'dt': datetime(2000, 1, 1), 'value': 23},
-            {'dt': datetime(2000, 1, 2), 'value': 47},
-            {'dt': datetime(2000, 1, 3), 'value': 71},
-            {'dt': datetime(2000, 1, 4), 'value': 95},
+            {'dt': datetime(2000, 1, 1), 'value': 23.0},
+            {'dt': datetime(2000, 1, 2), 'value': 47.0},
+            {'dt': datetime(2000, 1, 3), 'value': 71.0},
+            {'dt': datetime(2000, 1, 4), 'value': 95.0},
         ]))
 
     def test_resample_aggregation_import(self):
 
         out = resample(self.df.set_index('dt'), '1d', aggregation='numpy.max')
 
         assert_frame_equal(out, pd.DataFrame([
-            {'dt': datetime(2000, 1, 1), 'value': 23},
-            {'dt': datetime(2000, 1, 2), 'value': 47},
-            {'dt': datetime(2000, 1, 3), 'value': 71},
-            {'dt': datetime(2000, 1, 4), 'value': 95},
+            {'dt': datetime(2000, 1, 1), 'value': 23.0},
+            {'dt': datetime(2000, 1, 2), 'value': 47.0},
+            {'dt': datetime(2000, 1, 3), 'value': 71.0},
+            {'dt': datetime(2000, 1, 4), 'value': 95.0},
         ]))