Ravin Kohli: [ADD] documentation for pipelines and steps (#329)

automl · Nov 18, 2021 · 139d0ab · 139d0ab
1 parent a0d6e7c
commit 139d0ab
Show file tree

Hide file tree

Showing 39 changed files with 3,948 additions and 483 deletions.
diff --git a/development/_downloads/bc82bea3a5dd7bdba60b65220891d9e5/examples_python.zip b/development/_downloads/bc82bea3a5dd7bdba60b65220891d9e5/examples_python.zip
diff --git a/development/_downloads/fb625db3c50d423b1b7881136ffdeec8/examples_jupyter.zip b/development/_downloads/fb625db3c50d423b1b7881136ffdeec8/examples_jupyter.zip
diff --git a/development/_images/sphx_glr_example_visualization_001.png b/development/_images/sphx_glr_example_visualization_001.png
diff --git a/development/_images/sphx_glr_example_visualization_002.png b/development/_images/sphx_glr_example_visualization_002.png
diff --git a/development/_images/sphx_glr_example_visualization_thumb.png b/development/_images/sphx_glr_example_visualization_thumb.png
diff --git a/development/_modules/autoPyTorch/api/tabular_classification.html b/development/_modules/autoPyTorch/api/tabular_classification.html
@@ -261,6 +261,7 @@ <h1>Source code for autoPyTorch.api.tabular_classification</h1><div class="highl
 <span class="sd">        Fit both optimizes the machine learning models and builds an ensemble out of them.</span>
 <span class="sd">        To disable ensembling, set ensemble_size==0.</span>
 <span class="sd">        using the optimizer.</span>
+
 <span class="sd">        Args:</span>
 <span class="sd">            X_train, y_train, X_test, y_test: Union[np.ndarray, List, pd.DataFrame]</span>
 <span class="sd">                A pair of features (X_train) and targets (y_train) used to fit a</span>

diff --git a/development/_modules/autoPyTorch/api/tabular_regression.html b/development/_modules/autoPyTorch/api/tabular_regression.html
@@ -262,6 +262,7 @@ <h1>Source code for autoPyTorch.api.tabular_regression</h1><div class="highlight
 <span class="sd">        Fit both optimizes the machine learning models and builds an ensemble out of them.</span>
 <span class="sd">        To disable ensembling, set ensemble_size==0.</span>
 <span class="sd">        using the optimizer.</span>
+
 <span class="sd">        Args:</span>
 <span class="sd">            X_train, y_train, X_test, y_test: Union[np.ndarray, List, pd.DataFrame]</span>
 <span class="sd">                A pair of features (X_train) and targets (y_train) used to fit a</span>

diff --git a/development/_modules/autoPyTorch/pipeline/components/base_choice.html b/development/_modules/autoPyTorch/pipeline/components/base_choice.html
diff --git a/development/_modules/autoPyTorch/pipeline/components/base_component.html b/development/_modules/autoPyTorch/pipeline/components/base_component.html
diff --git a/development/_modules/autoPyTorch/pipeline/tabular_classification.html b/development/_modules/autoPyTorch/pipeline/tabular_classification.html
diff --git a/development/_modules/autoPyTorch/pipeline/tabular_regression.html b/development/_modules/autoPyTorch/pipeline/tabular_regression.html
diff --git a/development/_modules/autoPyTorch/pipeline/traditional_tabular_classification.html b/development/_modules/autoPyTorch/pipeline/traditional_tabular_classification.html
diff --git a/development/_modules/autoPyTorch/pipeline/traditional_tabular_regression.html b/development/_modules/autoPyTorch/pipeline/traditional_tabular_regression.html
diff --git a/development/_modules/index.html b/development/_modules/index.html
@@ -113,6 +113,12 @@
   <h1>All modules for which code is available</h1>
 <ul><li><a href="autoPyTorch/api/tabular_classification.html">autoPyTorch.api.tabular_classification</a></li>
 <li><a href="autoPyTorch/api/tabular_regression.html">autoPyTorch.api.tabular_regression</a></li>
+<li><a href="autoPyTorch/pipeline/components/base_choice.html">autoPyTorch.pipeline.components.base_choice</a></li>
+<li><a href="autoPyTorch/pipeline/components/base_component.html">autoPyTorch.pipeline.components.base_component</a></li>
+<li><a href="autoPyTorch/pipeline/tabular_classification.html">autoPyTorch.pipeline.tabular_classification</a></li>
+<li><a href="autoPyTorch/pipeline/tabular_regression.html">autoPyTorch.pipeline.tabular_regression</a></li>
+<li><a href="autoPyTorch/pipeline/traditional_tabular_classification.html">autoPyTorch.pipeline.traditional_tabular_classification</a></li>
+<li><a href="autoPyTorch/pipeline/traditional_tabular_regression.html">autoPyTorch.pipeline.traditional_tabular_regression</a></li>
 </ul>
 
     </div>

diff --git a/development/_sources/api.rst.txt b/development/_sources/api.rst.txt
@@ -23,4 +23,48 @@ Regression
 
 .. autoclass:: autoPyTorch.api.tabular_regression.TabularRegressionTask
     :members:
-    :inherited-members: search, refit, predict, score
+    :inherited-members: search, refit, predict, score
+
+
+=========
+Pipelines
+=========
+
+~~~~~~~~~~~~~~~~~~~~~~
+Tabular Classification
+~~~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: autoPyTorch.pipeline.tabular_classification.TabularClassificationPipeline
+    :members:
+
+.. autoclass:: autoPyTorch.pipeline.traditional_tabular_classification.TraditionalTabularClassificationPipeline
+    :members:
+
+~~~~~~~~~~~~~~~~~~
+Tabular Regression
+~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: autoPyTorch.pipeline.tabular_regression.TabularRegressionPipeline
+    :members:
+
+.. autoclass:: autoPyTorch.pipeline.traditional_tabular_regression.TraditionalTabularRegressionPipeline
+    :members:
+
+=================
+Steps in Pipeline
+=================
+
+
+~~~~~~~~~~~~~~~~~~~~
+autoPyTorchComponent
+~~~~~~~~~~~~~~~~~~~~
+
+.. autoclass:: autoPyTorch.pipeline.components.base_component.autoPyTorchComponent
+    :members:
+
+~~~~~~~~~~~~~~~~~
+autoPyTorchChoice
+~~~~~~~~~~~~~~~~~
+
+.. autoclass:: autoPyTorch.pipeline.components.base_choice.autoPyTorchChoice
+    :members:
diff --git a/development/_sources/examples/20_basics/example_image_classification.rst.txt b/development/_sources/examples/20_basics/example_image_classification.rst.txt
@@ -85,26 +85,25 @@ Image Classification
     Pipeline Random Config:
      ________________________________________ 
     Configuration:
-      image_augmenter:GaussianBlur:sigma_min, Value: 1.1477050650598004
-      image_augmenter:GaussianBlur:sigma_offset, Value: 0.2823191510883727
-      image_augmenter:GaussianBlur:use_augmenter, Value: True
+      image_augmenter:GaussianBlur:use_augmenter, Value: False
       image_augmenter:GaussianNoise:use_augmenter, Value: False
-      image_augmenter:RandomAffine:rotate, Value: 313
-      image_augmenter:RandomAffine:scale_offset, Value: 0.22157150483045607
-      image_augmenter:RandomAffine:shear, Value: 33
-      image_augmenter:RandomAffine:translate_percent_offset, Value: 0.31366795252179314
+      image_augmenter:RandomAffine:rotate, Value: 114
+      image_augmenter:RandomAffine:scale_offset, Value: 0.11914307425363649
+      image_augmenter:RandomAffine:shear, Value: 8
+      image_augmenter:RandomAffine:translate_percent_offset, Value: 0.027196925844249223
       image_augmenter:RandomAffine:use_augmenter, Value: True
-      image_augmenter:RandomCutout:use_augmenter, Value: False
-      image_augmenter:Resize:use_augmenter, Value: True
-      image_augmenter:ZeroPadAndCrop:percent, Value: 0.4994526394988469
-      normalizer:__choice__, Value: 'NoNormalizer'
+      image_augmenter:RandomCutout:p, Value: 0.24908507480752906
+      image_augmenter:RandomCutout:use_augmenter, Value: True
+      image_augmenter:Resize:use_augmenter, Value: False
+      image_augmenter:ZeroPadAndCrop:percent, Value: 0.2779993065586726
+      normalizer:__choice__, Value: 'ImageNormalizer'
 
     Fitting the pipeline...
     ________________________________________
             ImageClassificationPipeline
     ________________________________________
     0-) normalizer: 
-            NoNormalizer
+            ImageNormalizer
 
     1-) preprocessing: 
             EarlyPreprocessing
@@ -176,7 +175,7 @@ Image Classification
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 0 minutes  5.335 seconds)
+   **Total running time of the script:** ( 0 minutes  5.899 seconds)
 
 
 .. _sphx_glr_download_examples_20_basics_example_image_classification.py:

diff --git a/development/_sources/examples/20_basics/example_tabular_classification.rst.txt b/development/_sources/examples/20_basics/example_tabular_classification.rst.txt
@@ -133,7 +133,7 @@ Search for an ensemble of machine learning algorithms
  .. code-block:: none
 
 
-    <autoPyTorch.api.tabular_classification.TabularClassificationTask object at 0x7f9ce60aad00>
+    <autoPyTorch.api.tabular_classification.TabularClassificationTask object at 0x7f17d1cd15e0>
 
 
 
@@ -162,7 +162,7 @@ Print the final ensemble performance
 
  .. code-block:: none
 
-    <smac.runhistory.runhistory.RunHistory object at 0x7f9ce60aadc0> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
+    <smac.runhistory.runhistory.RunHistory object at 0x7f17d1d96e50> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
       data_loader:batch_size, Value: 64
       encoder:__choice__, Value: 'OneHotEncoder'
       feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -194,7 +194,7 @@ Print the final ensemble performance
       scaler:__choice__, Value: 'StandardScaler'
       trainer:StandardTrainer:weighted_loss, Value: True
       trainer:__choice__, Value: 'StandardTrainer'
-    , ta_runs=0, ta_time_used=0.0, wallclock_time=0.0012331008911132812, budget=0), TrajEntry(train_perf=0.1578947368421053, incumbent_id=1, incumbent=Configuration:
+    , ta_runs=0, ta_time_used=0.0, wallclock_time=0.0011916160583496094, budget=0), TrajEntry(train_perf=0.1578947368421053, incumbent_id=1, incumbent=Configuration:
       data_loader:batch_size, Value: 64
       encoder:__choice__, Value: 'OneHotEncoder'
       feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -226,59 +226,68 @@ Print the final ensemble performance
       scaler:__choice__, Value: 'StandardScaler'
       trainer:StandardTrainer:weighted_loss, Value: True
       trainer:__choice__, Value: 'StandardTrainer'
-    , ta_runs=1, ta_time_used=3.5240633487701416, wallclock_time=4.555788516998291, budget=5.555555555555555), TrajEntry(train_perf=0.14035087719298245, incumbent_id=2, incumbent=Configuration:
-      data_loader:batch_size, Value: 148
-      encoder:__choice__, Value: 'OneHotEncoder'
-      feature_preprocessor:TruncatedSVD:target_dim, Value: 5
-      feature_preprocessor:__choice__, Value: 'TruncatedSVD'
-      imputer:categorical_strategy, Value: 'constant_!missing!'
+    , ta_runs=1, ta_time_used=3.463505744934082, wallclock_time=4.493630886077881, budget=5.555555555555555), TrajEntry(train_perf=0.14035087719298245, incumbent_id=2, incumbent=Configuration:
+      data_loader:batch_size, Value: 170
+      encoder:__choice__, Value: 'NoEncoder'
+      feature_preprocessor:Nystroem:kernel, Value: 'cosine'
+      feature_preprocessor:Nystroem:n_components, Value: 5
+      feature_preprocessor:__choice__, Value: 'Nystroem'
+      imputer:categorical_strategy, Value: 'most_frequent'
       imputer:numerical_strategy, Value: 'constant_zero'
-      lr_scheduler:ExponentialLR:gamma, Value: 0.8253455276715344
-      lr_scheduler:__choice__, Value: 'ExponentialLR'
-      network_backbone:ShapedMLPBackbone:activation, Value: 'tanh'
-      network_backbone:ShapedMLPBackbone:max_units, Value: 950
-      network_backbone:ShapedMLPBackbone:mlp_shape, Value: 'long_funnel'
-      network_backbone:ShapedMLPBackbone:num_groups, Value: 12
-      network_backbone:ShapedMLPBackbone:output_dim, Value: 426
-      network_backbone:ShapedMLPBackbone:use_dropout, Value: False
+      lr_scheduler:__choice__, Value: 'NoScheduler'
+      network_backbone:ShapedMLPBackbone:activation, Value: 'sigmoid'
+      network_backbone:ShapedMLPBackbone:max_dropout, Value: 0.2058174477473529
+      network_backbone:ShapedMLPBackbone:max_units, Value: 656
+      network_backbone:ShapedMLPBackbone:mlp_shape, Value: 'stairs'
+      network_backbone:ShapedMLPBackbone:num_groups, Value: 7
+      network_backbone:ShapedMLPBackbone:output_dim, Value: 894
+      network_backbone:ShapedMLPBackbone:use_dropout, Value: True
       network_backbone:__choice__, Value: 'ShapedMLPBackbone'
       network_embedding:__choice__, Value: 'NoEmbedding'
       network_head:__choice__, Value: 'fully_connected'
       network_head:fully_connected:activation, Value: 'sigmoid'
       network_head:fully_connected:num_layers, Value: 2
-      network_head:fully_connected:units_layer_1, Value: 205
-      network_init:KaimingInit:bias_strategy, Value: 'Zero'
-      network_init:__choice__, Value: 'KaimingInit'
-      optimizer:AdamWOptimizer:beta1, Value: 0.8736175987868154
-      optimizer:AdamWOptimizer:beta2, Value: 0.9829288984771918
-      optimizer:AdamWOptimizer:lr, Value: 0.0004977933591642692
-      optimizer:AdamWOptimizer:weight_decay, Value: 0.09691900491056754
+      network_head:fully_connected:units_layer_1, Value: 218
+      network_init:NoInit:bias_strategy, Value: 'Normal'
+      network_init:__choice__, Value: 'NoInit'
+      optimizer:AdamWOptimizer:beta1, Value: 0.8602054373347771
+      optimizer:AdamWOptimizer:beta2, Value: 0.952272648589522
+      optimizer:AdamWOptimizer:lr, Value: 0.00038032217671957316
+      optimizer:AdamWOptimizer:weight_decay, Value: 0.05235035327160109
       optimizer:__choice__, Value: 'AdamWOptimizer'
-      scaler:__choice__, Value: 'MinMaxScaler'
-      trainer:MixUpTrainer:alpha, Value: 0.9512542558695131
+      scaler:Normalizer:norm, Value: 'mean_squared'
+      scaler:__choice__, Value: 'Normalizer'
+      trainer:MixUpTrainer:alpha, Value: 0.758019642405335
       trainer:MixUpTrainer:weighted_loss, Value: False
       trainer:__choice__, Value: 'MixUpTrainer'
-    , ta_runs=15, ta_time_used=121.69535493850708, wallclock_time=171.1178548336029, budget=50.0)]
-    {'accuracy': 0.8554913294797688}
-    |    | Preprocessing                                             | Estimator                                                       |   Weight |
-    |---:|:----------------------------------------------------------|:----------------------------------------------------------------|---------:|
-    |  0 | None                                                      | RFLearner                                                       |     0.24 |
-    |  1 | SimpleImputer,OneHotEncoder,Normalizer,KernelPCA          | embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.22 |
-    |  2 | None                                                      | SVMLearner                                                      |     0.18 |
-    |  3 | None                                                      | CBLearner                                                       |     0.1  |
-    |  4 | None                                                      | ETLearner                                                       |     0.08 |
-    |  5 | None                                                      | KNNLearner                                                      |     0.08 |
-    |  6 | SimpleImputer,OneHotEncoder,MinMaxScaler,PowerTransformer | embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.04 |
-    |  7 | SimpleImputer,NoEncoder,Normalizer,Nystroem               | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential |     0.04 |
-    |  8 | None                                                      | LGBMLearner                                                     |     0.02 |
+    , ta_runs=15, ta_time_used=109.08182263374329, wallclock_time=159.79244685173035, budget=50.0)]
+    {'accuracy': 0.861271676300578}
+    |    | Preprocessing                                                     | Estimator                                                          |   Weight |
+    |---:|:------------------------------------------------------------------|:-------------------------------------------------------------------|---------:|
+    |  0 | None                                                              | CBLearner                                                          |     0.18 |
+    |  1 | SimpleImputer,OneHotEncoder,Normalizer,KernelPCA                  | embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential       |     0.14 |
+    |  2 | SimpleImputer,OneHotEncoder,Normalizer,KitchenSink                | embedding,MLPBackbone,FullyConnectedHead,nn.Sequential             |     0.14 |
+    |  3 | SimpleImputer,NoEncoder,Normalizer,Nystroem                       | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.12 |
+    |  4 | SimpleImputer,NoEncoder,Normalizer,KitchenSink                    | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential          |     0.1  |
+    |  5 | SimpleImputer,NoEncoder,Normalizer,Nystroem                       | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.08 |
+    |  6 | SimpleImputer,OneHotEncoder,Normalizer,PowerTransformer           | embedding,ResNetBackbone,FullyConnectedHead,nn.Sequential          |     0.04 |
+    |  7 | SimpleImputer,NoEncoder,Normalizer,KernelPCA                      | no embedding,ShapedResNetBackbone,FullyConnectedHead,nn.Sequential |     0.04 |
+    |  8 | SimpleImputer,OneHotEncoder,MinMaxScaler,PowerTransformer         | embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential       |     0.02 |
+    |  9 | SimpleImputer,OneHotEncoder,MinMaxScaler,TruncSVD                 | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.02 |
+    | 10 | None                                                              | RFLearner                                                          |     0.02 |
+    | 11 | None                                                              | ETLearner                                                          |     0.02 |
+    | 12 | None                                                              | SVMLearner                                                         |     0.02 |
+    | 13 | None                                                              | KNNLearner                                                         |     0.02 |
+    | 14 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.02 |
+    | 15 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential    |     0.02 |
 
 
 
 
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 5 minutes  19.860 seconds)
+   **Total running time of the script:** ( 5 minutes  34.529 seconds)
 
 
 .. _sphx_glr_download_examples_20_basics_example_tabular_classification.py:

diff --git a/development/_sources/examples/20_basics/example_tabular_regression.rst.txt b/development/_sources/examples/20_basics/example_tabular_regression.rst.txt
@@ -125,7 +125,7 @@ Search for an ensemble of machine learning algorithms
  .. code-block:: none
 
 
-    <autoPyTorch.api.tabular_regression.TabularRegressionTask object at 0x7f9d88915a90>
+    <autoPyTorch.api.tabular_regression.TabularRegressionTask object at 0x7f1874c24ca0>
 
 
 
@@ -157,7 +157,7 @@ Print the final ensemble performance
 
  .. code-block:: none
 
-    <smac.runhistory.runhistory.RunHistory object at 0x7f9d88915e80> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
+    <smac.runhistory.runhistory.RunHistory object at 0x7f185d369790> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
       data_loader:batch_size, Value: 64
       encoder:__choice__, Value: 'OneHotEncoder'
       feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -188,7 +188,7 @@ Print the final ensemble performance
       optimizer:__choice__, Value: 'AdamOptimizer'
       scaler:__choice__, Value: 'StandardScaler'
       trainer:__choice__, Value: 'StandardTrainer'
-    , ta_runs=0, ta_time_used=0.0, wallclock_time=0.0011785030364990234, budget=0), TrajEntry(train_perf=0.3195903909944855, incumbent_id=1, incumbent=Configuration:
+    , ta_runs=0, ta_time_used=0.0, wallclock_time=0.0011248588562011719, budget=0), TrajEntry(train_perf=0.3195903909944855, incumbent_id=1, incumbent=Configuration:
       data_loader:batch_size, Value: 64
       encoder:__choice__, Value: 'OneHotEncoder'
       feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -219,7 +219,7 @@ Print the final ensemble performance
       optimizer:__choice__, Value: 'AdamOptimizer'
       scaler:__choice__, Value: 'StandardScaler'
       trainer:__choice__, Value: 'StandardTrainer'
-    , ta_runs=1, ta_time_used=2.167248249053955, wallclock_time=3.1977641582489014, budget=5.555555555555555)]
+    , ta_runs=1, ta_time_used=2.1191773414611816, wallclock_time=3.1469626426696777, budget=5.555555555555555)]
     {'r2': 0.9445248186059718}
     |    | Preprocessing                                                     | Estimator                                                       |   Weight |
     |---:|:------------------------------------------------------------------|:----------------------------------------------------------------|---------:|
@@ -234,7 +234,7 @@ Print the final ensemble performance
 
 .. rst-class:: sphx-glr-timing
 
-   **Total running time of the script:** ( 5 minutes  28.349 seconds)
+   **Total running time of the script:** ( 5 minutes  29.258 seconds)
 
 
 .. _sphx_glr_download_examples_20_basics_example_tabular_regression.py: