Merge pull request #76 from bjherger/bool

Taking out booleans

README.md

@@ -188,7 +188,7 @@ board games with my partner in Seattle.
 
 ### Development
 
- - Nothing here yet
+ - Boolean support deprecated. Boolean (bool) data type can be treated as a special case of categorical data types
 
 ### 2.0.2
 

keras_pandas/Automater.py

@@ -23,8 +23,8 @@ def __init__(self, numerical_vars=list(), categorical_vars=list(), boolean_vars=
         # Set up variable type dict, with entries <variable_type, list of variables>
         self._variable_type_dict = dict()
         self._variable_type_dict['numerical_vars'] = numerical_vars
-        self._variable_type_dict['categorical_vars'] = categorical_vars
-        self._variable_type_dict['boolean_vars'] = boolean_vars
+        # Categorical variables include both categorical and boolean
+        self._variable_type_dict['categorical_vars'] = categorical_vars + boolean_vars
         self._variable_type_dict['datetime_vars'] = datetime_vars
         self._variable_type_dict['text_vars'] = text_vars
         self._variable_type_dict['non_transformed_vars'] = non_transformed_vars

keras_pandas/constants.py

@@ -7,23 +7,21 @@
 from keras.layers import Embedding, Flatten, Bidirectional, LSTM
 from sklearn.preprocessing import Imputer, StandardScaler
 
-from keras_pandas.transformations import EmbeddingVectorizer, CategoricalImputer, LabelEncoder
+from keras_pandas.transformations import EmbeddingVectorizer, CategoricalImputer, LabelEncoder, StringEncoder
 
 default_sklearn_mapper_pipelines = defaultdict(lambda: list())
 
 default_sklearn_mapper_pipelines.update({
     'numerical_vars': [Imputer(strategy='mean'), StandardScaler()],
-    'categorical_vars': [CategoricalImputer(strategy='constant', fill_value='UNK', fill_unknown_labels=True),
+    'categorical_vars': [StringEncoder(), CategoricalImputer(strategy='constant', fill_value='UNK', fill_unknown_labels=True),
                          LabelEncoder()],
-    'boolean_vars': [LabelEncoder()],
-    'text_vars': [EmbeddingVectorizer()],
+    'text_vars': [StringEncoder(), EmbeddingVectorizer()],
     'non_transformed_vars': []
 })
 
 default_suggested_losses = {
     'numerical_vars': losses.mean_squared_error,
     'categorical_vars': losses.sparse_categorical_crossentropy,
-    'boolean_vars': losses.sparse_categorical_crossentropy
 }
 
 

keras_pandas/transformations.py

@@ -402,4 +402,15 @@ def inverse_transform(self, y):
         if diff:
             raise ValueError("y contains new labels: %s" % str(diff))
         y = numpy.asarray(y)
-        return self.classes_[y]
+        return self.classes_[y]
+
+class StringEncoder(BaseEstimator, TransformerMixin):
+
+    def __init__(self):
+        pass
+
+    def fit(self, X, y=None):
+        return self
+
+    def transform(self, X):
+        return X.astype(str)

tests/testautomater.py

@@ -84,7 +84,7 @@ def test_initializer(self):
         # Base case: No variables
         auto = Automater()
         self.assertEqual({'numerical_vars': list(), 'categorical_vars': list(),
-                          'boolean_vars': list(), 'datetime_vars': list(), 'text_vars': list(),
+                          'datetime_vars': list(), 'text_vars': list(),
                           'non_transformed_vars': list()}, auto._variable_type_dict, )
         self.assertCountEqual(list(), auto._user_provided_variables)
 
@@ -96,7 +96,6 @@ def test_initializer(self):
         }
 
         response = copy.deepcopy(data)
-        response['boolean_vars'] = list()
         response['non_transformed_vars'] = list()
         response['text_vars'] = list()
 
@@ -116,7 +115,6 @@ def test_initializer(self):
         }
 
         response = copy.deepcopy(data)
-        response['boolean_vars'] = list()
         response['non_transformed_vars'] = list()
 
         self.assertRaises(ValueError, Automater().__init__(), numerical_vars=data['numerical_vars'],

tests/testcategorical.py

@@ -103,6 +103,45 @@ def test_create_input_nub_numerical(self):
             _create_input_nub(variable_type_dict, train_df)
         self.assertEqual(3, len(input_layers))
 
+    def test_boolean(self):
+        observations = lib.load_mushroom()
+        observations['population_bool'] = observations['population'] == 's'
+
+        msk = numpy.random.rand(len(observations)) < 0.95
+        mushroom_train = observations[msk]
+        mushroom_test = observations[~msk]
+
+        categorical_vars = ['odor', 'habitat', 'class']
+        boolean_vars = ['population_bool']
+
+        auto = Automater(categorical_vars=categorical_vars, boolean_vars=boolean_vars,  response_var='class')
+
+        auto.fit(mushroom_train)
+        X_train, y_train = auto.transform(mushroom_train)
+
+        # Extract input_nub from auto
+        input_nub = auto.input_nub
+
+        # Extract output_nub from auto
+        output_nub = auto.output_nub
+
+        # Create DL model
+        x = input_nub
+        x = Dense(30)(x)
+        x = output_nub(x)
+
+        model = Model(inputs=auto.input_layers, outputs=x)
+        model.compile(optimizer='Adam', loss=auto.loss)
+
+        # Train DL model
+        model.fit(X_train, y_train)
+
+        # Transform test set
+        mushroom_test = mushroom_test.drop('class', axis=1)
+        X_test, y_test = auto.transform(mushroom_test)
+        model.predict(X_test)
+
+
     def test_categorical_whole(self):
         # St up data set
         mushroom_df = lib.load_mushroom()