Skip to content

Commit 5f1764a

Browse files
author
Github Actions
committed
Ravin Kohli: Hyperparameter Search Space updates now with constant and include ability (#146)
1 parent e30a075 commit 5f1764a

29 files changed

+393
-541
lines changed

‎refactor_development/_downloads/2f0f54a9831653ac5c12ee0e06324a62/basics_tabular_jupyter.zip‎

0 Bytes
Binary file not shown.

‎refactor_development/_downloads/8ef6602cf8ed40221edd89244fb031af/basics_tabular_python.zip‎

0 Bytes
Binary file not shown.

‎refactor_development/_downloads/b38295cf56f02b1c4547385a8f389f90/advanced_tabular_python.zip‎

0 Bytes
Binary file not shown.

‎refactor_development/_downloads/df038041811db9bd567de83692a8f994/advanced_tabular_jupyter.zip‎

0 Bytes
Binary file not shown.

‎refactor_development/_modules/autoPyTorch/api/tabular_classification.html‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -407,7 +407,7 @@ <h1>Source code for autoPyTorch.api.tabular_classification</h1><div class="highl
407407
</p>
408408
<p>
409409
&copy; Copyright 2014-2019, Machine Learning Professorship Freiburg.<br/>
410-
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 3.5.2.<br/>
410+
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 3.5.3.<br/>
411411
</p>
412412
</div>
413413
</footer>

‎refactor_development/_modules/index.html‎

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -128,7 +128,7 @@ <h1>All modules for which code is available</h1>
128128
</p>
129129
<p>
130130
&copy; Copyright 2014-2019, Machine Learning Professorship Freiburg.<br/>
131-
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 3.5.2.<br/>
131+
Created using <a href="http://sphinx-doc.org/">Sphinx</a> 3.5.3.<br/>
132132
</p>
133133
</div>
134134
</footer>

‎refactor_development/_sources/advanced_tabular/example_custom_configuration_space.rst.txt‎

Lines changed: 66 additions & 64 deletions
Original file line numberDiff line numberDiff line change
@@ -46,7 +46,7 @@ the search. Currently, there are two changes that can be made to the space:-
4646

4747
.. code-block:: none
4848
49-
<smac.runhistory.runhistory.RunHistory object at 0x7f9ce8692b50> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
49+
<smac.runhistory.runhistory.RunHistory object at 0x7ff8fe8a32e0> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
5050
data_loader:batch_size, Value: 32
5151
encoder:__choice__, Value: 'OneHotEncoder'
5252
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -65,16 +65,7 @@ the search. Currently, there are two changes that can be made to the space:-
6565
network_backbone:MLPBackbone:num_units_5, Value: 200
6666
network_backbone:MLPBackbone:use_dropout, Value: False
6767
network_backbone:__choice__, Value: 'MLPBackbone'
68-
network_embedding:LearnedEntityEmbedding:dimension_reduction_0, Value: 0.5
69-
network_embedding:LearnedEntityEmbedding:dimension_reduction_1, Value: 0.5
70-
network_embedding:LearnedEntityEmbedding:dimension_reduction_2, Value: 0.5
71-
network_embedding:LearnedEntityEmbedding:dimension_reduction_3, Value: 0.5
72-
network_embedding:LearnedEntityEmbedding:dimension_reduction_4, Value: 0.5
73-
network_embedding:LearnedEntityEmbedding:dimension_reduction_5, Value: 0.5
74-
network_embedding:LearnedEntityEmbedding:dimension_reduction_6, Value: 0.5
75-
network_embedding:LearnedEntityEmbedding:dimension_reduction_7, Value: 0.5
76-
network_embedding:LearnedEntityEmbedding:min_unique_values_for_embedding, Value: 5
77-
network_embedding:__choice__, Value: 'LearnedEntityEmbedding'
68+
network_embedding:__choice__, Value: 'NoEmbedding'
7869
network_head:__choice__, Value: 'fully_connected'
7970
network_head:fully_connected:activation, Value: 'relu'
8071
network_head:fully_connected:num_layers, Value: 2
@@ -89,7 +80,7 @@ the search. Currently, there are two changes that can be made to the space:-
8980
scaler:__choice__, Value: 'StandardScaler'
9081
trainer:StandardTrainer:weighted_loss, Value: True
9182
trainer:__choice__, Value: 'StandardTrainer'
92-
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001926422119140625, budget=0), TrajEntry(train_perf=0.18128654970760238, incumbent_id=1, incumbent=Configuration:
83+
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.0020105838775634766, budget=0), TrajEntry(train_perf=0.16374269005847952, incumbent_id=1, incumbent=Configuration:
9384
data_loader:batch_size, Value: 32
9485
encoder:__choice__, Value: 'OneHotEncoder'
9586
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -108,16 +99,7 @@ the search. Currently, there are two changes that can be made to the space:-
10899
network_backbone:MLPBackbone:num_units_5, Value: 200
109100
network_backbone:MLPBackbone:use_dropout, Value: False
110101
network_backbone:__choice__, Value: 'MLPBackbone'
111-
network_embedding:LearnedEntityEmbedding:dimension_reduction_0, Value: 0.5
112-
network_embedding:LearnedEntityEmbedding:dimension_reduction_1, Value: 0.5
113-
network_embedding:LearnedEntityEmbedding:dimension_reduction_2, Value: 0.5
114-
network_embedding:LearnedEntityEmbedding:dimension_reduction_3, Value: 0.5
115-
network_embedding:LearnedEntityEmbedding:dimension_reduction_4, Value: 0.5
116-
network_embedding:LearnedEntityEmbedding:dimension_reduction_5, Value: 0.5
117-
network_embedding:LearnedEntityEmbedding:dimension_reduction_6, Value: 0.5
118-
network_embedding:LearnedEntityEmbedding:dimension_reduction_7, Value: 0.5
119-
network_embedding:LearnedEntityEmbedding:min_unique_values_for_embedding, Value: 5
120-
network_embedding:__choice__, Value: 'LearnedEntityEmbedding'
102+
network_embedding:__choice__, Value: 'NoEmbedding'
121103
network_head:__choice__, Value: 'fully_connected'
122104
network_head:fully_connected:activation, Value: 'relu'
123105
network_head:fully_connected:num_layers, Value: 2
@@ -132,50 +114,70 @@ the search. Currently, there are two changes that can be made to the space:-
132114
scaler:__choice__, Value: 'StandardScaler'
133115
trainer:StandardTrainer:weighted_loss, Value: True
134116
trainer:__choice__, Value: 'StandardTrainer'
135-
, ta_runs=1, ta_time_used=4.341261148452759, wallclock_time=5.757386207580566, budget=5.555555555555555), TrajEntry(train_perf=0.1578947368421053, incumbent_id=2, incumbent=Configuration:
136-
data_loader:batch_size, Value: 390
117+
, ta_runs=1, ta_time_used=5.319528579711914, wallclock_time=6.774984359741211, budget=5.555555555555555), TrajEntry(train_perf=0.15204678362573099, incumbent_id=2, incumbent=Configuration:
118+
data_loader:batch_size, Value: 475
137119
encoder:__choice__, Value: 'OneHotEncoder'
138120
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
139121
imputer:categorical_strategy, Value: 'most_frequent'
140-
imputer:numerical_strategy, Value: 'most_frequent'
141-
lr_scheduler:CosineAnnealingLR:T_max, Value: 60
142-
lr_scheduler:__choice__, Value: 'CosineAnnealingLR'
122+
imputer:numerical_strategy, Value: 'constant_zero'
123+
lr_scheduler:__choice__, Value: 'NoScheduler'
143124
network_backbone:MLPBackbone:activation, Value: 'tanh'
144-
network_backbone:MLPBackbone:dropout_1, Value: 0.3637598891246192
145-
network_backbone:MLPBackbone:dropout_2, Value: 0.06650911315696684
146-
network_backbone:MLPBackbone:num_groups, Value: 2
147-
network_backbone:MLPBackbone:num_units_1, Value: 546
148-
network_backbone:MLPBackbone:num_units_2, Value: 969
125+
network_backbone:MLPBackbone:dropout_1, Value: 0.579891279191762
126+
network_backbone:MLPBackbone:dropout_2, Value: 0.43202885747368863
127+
network_backbone:MLPBackbone:dropout_3, Value: 0.2053050533304992
128+
network_backbone:MLPBackbone:dropout_4, Value: 0.3628626567848122
129+
network_backbone:MLPBackbone:dropout_5, Value: 0.000687232634536894
130+
network_backbone:MLPBackbone:dropout_6, Value: 0.30779918180581656
131+
network_backbone:MLPBackbone:dropout_7, Value: 0.4566654226669556
132+
network_backbone:MLPBackbone:num_groups, Value: 7
133+
network_backbone:MLPBackbone:num_units_1, Value: 749
134+
network_backbone:MLPBackbone:num_units_2, Value: 751
135+
network_backbone:MLPBackbone:num_units_3, Value: 759
136+
network_backbone:MLPBackbone:num_units_4, Value: 664
137+
network_backbone:MLPBackbone:num_units_5, Value: 219
138+
network_backbone:MLPBackbone:num_units_6, Value: 757
139+
network_backbone:MLPBackbone:num_units_7, Value: 1005
149140
network_backbone:MLPBackbone:use_dropout, Value: True
150141
network_backbone:__choice__, Value: 'MLPBackbone'
151-
network_embedding:__choice__, Value: 'NoEmbedding'
142+
network_embedding:LearnedEntityEmbedding:dimension_reduction_0, Value: 0.9640640623783606
143+
network_embedding:LearnedEntityEmbedding:dimension_reduction_1, Value: 0.017233504391813814
144+
network_embedding:LearnedEntityEmbedding:dimension_reduction_2, Value: 0.24122690885917664
145+
network_embedding:LearnedEntityEmbedding:dimension_reduction_3, Value: 0.31247176333246596
146+
network_embedding:LearnedEntityEmbedding:dimension_reduction_4, Value: 0.41504826813841933
147+
network_embedding:LearnedEntityEmbedding:dimension_reduction_5, Value: 0.8395119637200936
148+
network_embedding:LearnedEntityEmbedding:dimension_reduction_6, Value: 0.8208414027523236
149+
network_embedding:LearnedEntityEmbedding:dimension_reduction_7, Value: 0.4284420622613293
150+
network_embedding:LearnedEntityEmbedding:min_unique_values_for_embedding, Value: 7
151+
network_embedding:__choice__, Value: 'LearnedEntityEmbedding'
152152
network_head:__choice__, Value: 'fully_connected'
153-
network_head:fully_connected:activation, Value: 'sigmoid'
154-
network_head:fully_connected:num_layers, Value: 2
155-
network_head:fully_connected:units_layer_1, Value: 172
153+
network_head:fully_connected:num_layers, Value: 1
156154
network_init:KaimingInit:bias_strategy, Value: 'Zero'
157155
network_init:__choice__, Value: 'KaimingInit'
158-
optimizer:AdamOptimizer:beta1, Value: 0.8884290707202807
159-
optimizer:AdamOptimizer:beta2, Value: 0.9029615465570149
160-
optimizer:AdamOptimizer:lr, Value: 0.026058153470738626
161-
optimizer:AdamOptimizer:weight_decay, Value: 0.09667386116218406
156+
optimizer:AdamOptimizer:beta1, Value: 0.9770847327434384
157+
optimizer:AdamOptimizer:beta2, Value: 0.9710627513919582
158+
optimizer:AdamOptimizer:lr, Value: 0.00010844892447274338
159+
optimizer:AdamOptimizer:weight_decay, Value: 0.05048412416506887
162160
optimizer:__choice__, Value: 'AdamOptimizer'
163-
scaler:__choice__, Value: 'MinMaxScaler'
164-
trainer:MixUpTrainer:alpha, Value: 0.02729109397127838
165-
trainer:MixUpTrainer:weighted_loss, Value: True
166-
trainer:__choice__, Value: 'MixUpTrainer'
167-
, ta_runs=17, ta_time_used=197.08009243011475, wallclock_time=238.20992231369019, budget=50.0)]
168-
{'accuracy': 0.8670520231213873}
169-
| | Preprocessing | Estimator | Weight |
170-
|---:|:----------------------------------------------------------------|:----------------------------------------------------------|---------:|
171-
| 0 | SimpleImputer,OneHotEncoder,Normalizer,KernelPCA | embedding,ResNetBackbone,FullyConnectedHead,nn.Sequential | 0.2 |
172-
| 1 | SimpleImputer,OneHotEncoder,MinMaxScaler,NoFeaturePreprocessing | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.2 |
173-
| 2 | None | CatBoostClassifier | 0.2 |
174-
| 3 | None | ExtraTreesClassifier | 0.18 |
175-
| 4 | SimpleImputer,OneHotEncoder,MinMaxScaler,NoFeaturePreprocessing | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.1 |
176-
| 5 | None | KNNClassifier | 0.1 |
177-
| 6 | SimpleImputer,OneHotEncoder,Normalizer,KernelPCA | embedding,ResNetBackbone,FullyConnectedHead,nn.Sequential | 0.02 |
178-
<smac.runhistory.runhistory.RunHistory object at 0x7f9cd57a5790> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
161+
scaler:Normalizer:norm, Value: 'max'
162+
scaler:__choice__, Value: 'Normalizer'
163+
trainer:StandardTrainer:weighted_loss, Value: False
164+
trainer:__choice__, Value: 'StandardTrainer'
165+
, ta_runs=11, ta_time_used=94.9950180053711, wallclock_time=118.93319940567017, budget=16.666666666666664)]
166+
{'accuracy': 0.8728323699421965}
167+
| | Preprocessing | Estimator | Weight |
168+
|---:|:------------------------------------------------------------------|:-------------------------------------------------------------|---------:|
169+
| 0 | None | CatBoostClassifier | 0.3 |
170+
| 1 | None | ExtraTreesClassifier | 0.2 |
171+
| 2 | SimpleImputer,OneHotEncoder,Normalizer,KernelPCA | embedding,ResNetBackbone,FullyConnectedHead,nn.Sequential | 0.1 |
172+
| 3 | SimpleImputer,OneHotEncoder,Normalizer,NoFeaturePreprocessing | embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
173+
| 4 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
174+
| 5 | None | KNNClassifier | 0.06 |
175+
| 6 | SimpleImputer,OneHotEncoder,StandardScaler,PowerTransformer | no embedding,ResNetBackbone,FullyConnectedHead,nn.Sequential | 0.04 |
176+
| 7 | SimpleImputer,OneHotEncoder,Normalizer,NoFeaturePreprocessing | embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.04 |
177+
| 8 | None | RFClassifier | 0.04 |
178+
| 9 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.04 |
179+
| 10 | None | LGBMClassifier | 0.02 |
180+
<smac.runhistory.runhistory.RunHistory object at 0x7ff8fc230d60> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
179181
data_loader:batch_size, Value: 32
180182
encoder:__choice__, Value: 'NoEncoder'
181183
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -207,7 +209,7 @@ the search. Currently, there are two changes that can be made to the space:-
207209
scaler:__choice__, Value: 'StandardScaler'
208210
trainer:StandardTrainer:weighted_loss, Value: True
209211
trainer:__choice__, Value: 'StandardTrainer'
210-
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.0015478134155273438, budget=0), TrajEntry(train_perf=0.21052631578947367, incumbent_id=1, incumbent=Configuration:
212+
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001617431640625, budget=0), TrajEntry(train_perf=0.21052631578947367, incumbent_id=1, incumbent=Configuration:
211213
data_loader:batch_size, Value: 32
212214
encoder:__choice__, Value: 'NoEncoder'
213215
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
@@ -239,16 +241,16 @@ the search. Currently, there are two changes that can be made to the space:-
239241
scaler:__choice__, Value: 'StandardScaler'
240242
trainer:StandardTrainer:weighted_loss, Value: True
241243
trainer:__choice__, Value: 'StandardTrainer'
242-
, ta_runs=1, ta_time_used=3.674795389175415, wallclock_time=5.081493616104126, budget=5.555555555555555)]
244+
, ta_runs=1, ta_time_used=4.8831260204315186, wallclock_time=6.359606027603149, budget=5.555555555555555)]
243245
{'accuracy': 0.8670520231213873}
244246
| | Preprocessing | Estimator | Weight |
245247
|---:|:----------------|:---------------------|---------:|
246-
| 0 | None | CatBoostClassifier | 0.28 |
247-
| 1 | None | RFClassifier | 0.24 |
248-
| 2 | None | ExtraTreesClassifier | 0.2 |
248+
| 0 | None | ExtraTreesClassifier | 0.28 |
249+
| 1 | None | CatBoostClassifier | 0.22 |
250+
| 2 | None | RFClassifier | 0.22 |
249251
| 3 | None | KNNClassifier | 0.14 |
250-
| 4 | None | LGBMClassifier | 0.08 |
251-
| 5 | None | SVC | 0.06 |
252+
| 4 | None | LGBMClassifier | 0.1 |
253+
| 5 | None | SVC | 0.04 |
252254
253255
254256
@@ -377,7 +379,7 @@ the search. Currently, there are two changes that can be made to the space:-
377379
378380
.. rst-class:: sphx-glr-timing
379381

380-
**Total running time of the script:** ( 11 minutes 23.476 seconds)
382+
**Total running time of the script:** ( 11 minutes 42.274 seconds)
381383

382384

383385
.. _sphx_glr_download_advanced_tabular_example_custom_configuration_space.py:

0 commit comments

Comments
 (0)