Docs (#938)

* docs * docs * format * lint
keras-team · Jan 31, 2020 · bfaa475 · bfaa475
1 parent 9f33532
commit bfaa475
Show file tree

Hide file tree

Showing 8 changed files with 37 additions and 41 deletions.
diff --git a/autokeras/hypermodels/wrapper.py b/autokeras/hypermodels/wrapper.py
@@ -111,19 +111,18 @@ class StructuredDataBlock(block_module.Block):
     """Block for structured data.
 
     # Arguments
-        feature_encoding: Boolean. Whether to use feature encoding block to encode
-            the categorical features. Defaults to True. If specified as None, it will
-            be tuned automatically.
+        categorical_encoding: Boolean. Whether to use the CategoricalToNumerical to
+            encode the categorical features to numerical features. Defaults to True.
+            If specified as None, it will be tuned automatically.
         seed: Int. Random seed.
     """
 
     def __init__(self,
-                 feature_encoding=True,
-                 block_type=None,
+                 categorical_encoding=True,
                  seed=None,
                  **kwargs):
         super().__init__(**kwargs)
-        self.feature_encoding = feature_encoding
+        self.categorical_encoding = categorical_encoding
         self.seed = seed
         self.column_types = None
         self.column_names = None
@@ -136,12 +135,12 @@ def get_config(self):
 
     def build_feature_encoding(self, hp, input_node):
         output_node = input_node
-        feature_encoding = self.feature_encoding
-        if feature_encoding is None:
-            feature_encoding = hp.Choice('feature_encoding',
-                                         [True, False],
-                                         default=True)
-        if feature_encoding:
+        categorical_encoding = self.categorical_encoding
+        if categorical_encoding is None:
+            categorical_encoding = hp.Choice('feature_encoding',
+                                             [True, False],
+                                             default=True)
+        if categorical_encoding:
             block = preprocessing.CategoricalToNumerical()
             block.column_types = self.column_types
             block.column_names = self.column_names

diff --git a/docs/autogen.py b/docs/autogen.py
@@ -55,7 +55,6 @@
     ],
     'base.md': [
         'autokeras.Node',
-        'autokeras.Preprocessor',
         'autokeras.Block',
         'autokeras.Block.build',
         'autokeras.Head',
@@ -69,7 +68,7 @@
     'block.md': [
         'autokeras.ConvBlock',
         'autokeras.DenseBlock',
-        'autokeras.EmbeddingBlock',
+        'autokeras.Embedding',
         'autokeras.Merge',
         'autokeras.ResNetBlock',
         'autokeras.RNNBlock',
@@ -81,12 +80,11 @@
         'autokeras.TextBlock',
     ],
     'preprocessor.md': [
-        'autokeras.FeatureEngineering',
         'autokeras.ImageAugmentation',
-        'autokeras.LightGBM',
         'autokeras.Normalization',
         'autokeras.TextToIntSequence',
         'autokeras.TextToNgramVector',
+        'autokeras.CategoricalToNumerical',
     ],
     'head.md': [
         'autokeras.ClassificationHead',

diff --git a/docs/templates/tutorial/customized.md b/docs/templates/tutorial/customized.md
@@ -151,7 +151,7 @@ print(auto_model.evaluate(x_test, y_test))
 **Blocks**:
 [ConvBlock](/block/#convblock-class),
 [DenseBlock](/block/#denseblock-class),
-[EmbeddingBlock](/block/#embeddingblock-class),
+[Embedding](/block/#embedding-class),
 [Merge](/block/#merge-class),
 [ResNetBlock](/block/#resnetblock-class),
 [RNNBlock](/block/#rnnblock-class),

diff --git a/docs/templates/tutorial/multi.md b/docs/templates/tutorial/multi.md
@@ -110,8 +110,8 @@ graph LR
     id3 --> id5(ResNet V2)
     id4 --> id6(Merge)
     id5 --> id6
-    id7(StructuredDataInput) --> id8(Feature Engineering)
-    id8 --> id9(LightGBM)
+    id7(StructuredDataInput) --> id8(CategoricalToNumerical)
+    id8 --> id9(DenseBlock)
     id6 --> id10(Merge)
     id9 --> id10
     id10 --> id11(Classification Head)
@@ -129,8 +129,8 @@ output_node2 = ak.ResNetBlock(version='v2')(output_node)
 output_node1 = ak.Merge()([output_node1, output_node2])
 
 input_node2 = ak.StructuredDataInput()
-output_node = ak.FeatureEngineering()(input_node2)
-output_node2 = ak.LightGBM()(output_node)
+output_node = ak.CategoricalToNumerical()(input_node2)
+output_node2 = ak.DenseBlock()(output_node)
 
 output_node = ak.Merge()([output_node1, output_node2])
 output_node1 = ak.ClassificationHead()(output_node)
@@ -160,5 +160,7 @@ You can also refer to the Data Format section of the tutorials of
 [AutoModel](/auto_model/#automodel-class),
 [ImageInput](/node/#imageinput-class),
 [StructuredDataInput](/node/#structureddatainput-class),
-[RegressionHead](/head/#regressionhead-class).
-[ClassificationHead](/head/#classificationhead-class).
+[DenseBlock](/block/#denseblock-class),
+[RegressionHead](/head/#regressionhead-class),
+[ClassificationHead](/head/#classificationhead-class),
+[CategoricalToNumerical](/preprocessor/#categoricaltonumerical-class).
diff --git a/docs/templates/tutorial/overview.md b/docs/templates/tutorial/overview.md
@@ -39,17 +39,16 @@ The following are the links to the documentation of the predefined input nodes a
 [TextInput](/node/#textinput-class).
 
 **Preprocessors**:
-[FeatureEngineering](/preprocessor/#featureengineering-class),
 [ImageAugmentation](/preprocessor/#imageaugmentation-class),
-[LightGBM](/preprocessor/#lightgbm-class),
 [Normalization](/preprocessor/#normalization-class),
 [TextToIntSequence](/preprocessor/#texttointsequence-class),
-[TextToNgramVector](/preprocessor/#texttongramvector-class).
+[TextToNgramVector](/preprocessor/#texttongramvector-class),
+[CategoricalToNumerical](/preprocessor/#categoricaltonumerical-class).
 
 **Blocks**:
 [ConvBlock](/block/#convblock-class),
 [DenseBlock](/block/#denseblock-class),
-[EmbeddingBlock](/block/#embeddingblock-class),
+[Embedding](/block/#embedding-class),
 [Merge](/block/#merge-class),
 [ResNetBlock](/block/#resnetblock-class),
 [RNNBlock](/block/#rnnblock-class),

diff --git a/docs/templates/tutorial/structured_data_classification.md b/docs/templates/tutorial/structured_data_classification.md
@@ -148,11 +148,8 @@ For advanced users, you may customize your search space by using
 [AutoModel](/auto_model/#automodel-class) instead of
 [StructuredDataClassifier](/structured_data_classifier). You can configure the
 [StructuredDataBlock](/block/#structureddatablock-class) for some high-level
-configurations, e.g., `feature_engineering` for whether to use the
-[FeatureEngineering](/preprocessor/#featureengineering-class) block, `block_type` for
-which type of block you want to use in the search space.  You can use 'dense' for
-[DenseBlock](/block/#denseblock-class), or you can use 'lightgbm' for
-[LightGBM](/preprocessor/#lightgbm-class).  You can also do not specify these
+configurations, e.g., `categorical_encoding` for whether to use the
+[CategoricalToNumerical](/preprocessor/#categoricaltonumerical-class). You can also do not specify these
 arguments, which would leave the different choices to be tuned automatically. See
 the following example for detail.
 
@@ -161,7 +158,7 @@ import autokeras as ak
 
 input_node = ak.StructuredDataInput()
 output_node = ak.StructuredDataBlock(
-    feature_engineering=False,
+    categorical_encoding=True,
     block_type='dense')(input_node)
 output_node = ak.ClassificationHead()(output_node)
 clf = ak.AutoModel(inputs=input_node, outputs=output_node, max_trials=10)
@@ -180,8 +177,8 @@ further. See the following example.
 import autokeras as ak
 
 input_node = ak.StructuredDataInput()
-output_node = ak.FeatureEngineering(max_columns=500)(input_node)
-output_node = ak.LightGBM()(output_node)
+output_node = ak.CategoricalToNumerical()(input_node)
+output_node = ak.DenseBlock()(output_node)
 output_node = ak.ClassificationHead()(output_node)
 clf = ak.AutoModel(inputs=input_node, outputs=output_node, max_trials=10)
 clf.fit(x_train, y_train)
@@ -193,8 +190,7 @@ clf.fit(x_train, y_train)
 [AutoModel](/auto_model/#automodel-class),
 [StructuredDataClassifier](/structured_data_classifier),
 [StructuredDataBlock](/block/#structureddatablock-class),
-[FeatureEngineering](/preprocessor/#featureengineering-class),
 [DenseBlock](/block/#denseblock-class),
-[LightGBM](/preprocessor/#lightgbm-class),
 [StructuredDataInput](/node/#structureddatainput-class),
-[ClassificationHead](/head/#classificationhead-class).
+[ClassificationHead](/head/#classificationhead-class),
+[CategoricalToNumerical](/preprocessor/#categoricaltonumerical-class).
diff --git a/docs/templates/tutorial/text_classification.md b/docs/templates/tutorial/text_classification.md
@@ -82,7 +82,7 @@ For advanced users, you may customize your search space by using
 [TextBlock](/block/#textblock-class) for some high-level configurations, e.g., `vectorizer`
 for the type of text vectorization method to use.  You can use 'sequence', which uses
 [TextToInteSequence](/preprocessor/#texttointsequence-class) to convert the words to
-integers and use [EmbeddingBlock](/block/#embeddingblock-class) for embedding the
+integers and use [Embedding](/block/#embedding-class) for embedding the
 integer sequences, or you can use 'ngram', which uses
 [TextToNgramVector](/preprocessor/#texttongramvector-class) to vectorize the
 sentences.  You can also do not specify these arguments, which would leave the
@@ -111,7 +111,7 @@ import autokeras as ak
 
 input_node = ak.TextInput()
 output_node = ak.TextToIntSequence()(input_node)
-output_node = ak.EmbeddingBlock()(output_node)
+output_node = ak.Embedding()(output_node)
 # Use separable Conv layers in Keras.
 output_node = ak.ConvBlock(separable=True)(output_node)
 output_node = ak.ClassificationHead()(output_node)
@@ -152,7 +152,7 @@ print(clf.evaluate(test_set))
 [AutoModel](/auto_model/#automodel-class),
 [TextBlock](/block/#textblock-class),
 [TextToInteSequence](/preprocessor/#texttointsequence-class),
-[EmbeddingBlock](/block/#embeddingblock-class),
+[Embedding](/block/#embedding-class),
 [TextToNgramVector](/preprocessor/#texttongramvector-class),
 [ConvBlock](/block/#convblock-class),
 [TextInput](/node/#textinput-class),

diff --git a/shell/lint.sh b/shell/lint.sh
@@ -0,0 +1,2 @@
+isort -rc -sl -c
+flake8