Skip to content

Commit

Permalink
Ravin Kohli: [FIX] Additional metrics during training (#316)
Browse files Browse the repository at this point in the history
  • Loading branch information
Github Actions committed Nov 12, 2021
1 parent 1f28082 commit 8f70383
Show file tree
Hide file tree
Showing 23 changed files with 175 additions and 177 deletions.
Binary file not shown.
Binary file not shown.
Binary file modified development/_images/sphx_glr_example_visualization_001.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Binary file modified development/_images/sphx_glr_example_visualization_thumb.png
Sorry, something went wrong. Reload?
Sorry, we cannot display this file.
Sorry, this file is invalid so it cannot be displayed.
Original file line number Diff line number Diff line change
Expand Up @@ -85,19 +85,13 @@ Image Classification
Pipeline Random Config:
________________________________________
Configuration:
image_augmenter:GaussianBlur:sigma_min, Value: 0.03894046691345188
image_augmenter:GaussianBlur:sigma_offset, Value: 1.3348068045107708
image_augmenter:GaussianBlur:use_augmenter, Value: True
image_augmenter:GaussianNoise:use_augmenter, Value: False
image_augmenter:RandomAffine:rotate, Value: 334
image_augmenter:RandomAffine:scale_offset, Value: 0.37914615264461815
image_augmenter:RandomAffine:shear, Value: 1
image_augmenter:RandomAffine:translate_percent_offset, Value: 0.13915163483786533
image_augmenter:RandomAffine:use_augmenter, Value: True
image_augmenter:RandomCutout:p, Value: 0.9948220651231314
image_augmenter:RandomCutout:use_augmenter, Value: True
image_augmenter:GaussianBlur:use_augmenter, Value: False
image_augmenter:GaussianNoise:sigma_offset, Value: 0.43287833896879524
image_augmenter:GaussianNoise:use_augmenter, Value: True
image_augmenter:RandomAffine:use_augmenter, Value: False
image_augmenter:RandomCutout:use_augmenter, Value: False
image_augmenter:Resize:use_augmenter, Value: False
image_augmenter:ZeroPadAndCrop:percent, Value: 0.43053137159256827
image_augmenter:ZeroPadAndCrop:percent, Value: 0.4972822105141404
normalizer:__choice__, Value: 'ImageNormalizer'

Fitting the pipeline...
Expand Down Expand Up @@ -177,7 +171,7 @@ Image Classification
.. rst-class:: sphx-glr-timing

**Total running time of the script:** ( 0 minutes 6.162 seconds)
**Total running time of the script:** ( 0 minutes 5.855 seconds)


.. _sphx_glr_download_examples_20_basics_example_image_classification.py:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -133,7 +133,7 @@ Search for an ensemble of machine learning algorithms
.. code-block:: none
<autoPyTorch.api.tabular_classification.TabularClassificationTask object at 0x7fe0faf6af40>
<autoPyTorch.api.tabular_classification.TabularClassificationTask object at 0x7ff1d130deb0>
Expand Down Expand Up @@ -162,7 +162,7 @@ Print the final ensemble performance

.. code-block:: none
<smac.runhistory.runhistory.RunHistory object at 0x7fe06e273730> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
<smac.runhistory.runhistory.RunHistory object at 0x7ff1d130d790> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 64
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -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=0.0014061927795410156, budget=0), TrajEntry(train_perf=0.16374269005847952, incumbent_id=1, incumbent=Configuration:
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001264810562133789, budget=0), TrajEntry(train_perf=0.16374269005847952, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 64
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -226,7 +226,7 @@ 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=3.995302677154541, wallclock_time=5.0308003425598145, budget=5.555555555555555), TrajEntry(train_perf=0.14035087719298245, incumbent_id=2, incumbent=Configuration:
, ta_runs=1, ta_time_used=3.4787635803222656, wallclock_time=4.510373592376709, budget=5.555555555555555), TrajEntry(train_perf=0.14035087719298245, incumbent_id=2, incumbent=Configuration:
data_loader:batch_size, Value: 266
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -270,28 +270,30 @@ Print the final ensemble performance
trainer:MixUpTrainer:alpha, Value: 0.3593787058162444
trainer:MixUpTrainer:weighted_loss, Value: False
trainer:__choice__, Value: 'MixUpTrainer'
, ta_runs=16, ta_time_used=140.4636766910553, wallclock_time=196.30338406562805, budget=50.0)]
{'accuracy': 0.8728323699421965}
, ta_runs=16, ta_time_used=129.56639432907104, wallclock_time=178.3990023136139, budget=50.0)]
{'accuracy': 0.8786127167630058}
| | Preprocessing | Estimator | Weight |
|---:|:------------------------------------------------------------------|:-------------------------------------------------------------------|---------:|
| 0 | None | CBLearner | 0.4 |
| 1 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential | 0.1 |
| 2 | SimpleImputer,OneHotEncoder,MinMaxScaler,NoFeaturePreprocessing | embedding,ShapedResNetBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
| 3 | SimpleImputer,OneHotEncoder,NoScaler,KernelPCA | no embedding,ShapedResNetBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
| 4 | None | RFLearner | 0.08 |
| 5 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
| 6 | None | ETLearner | 0.06 |
| 0 | SimpleImputer,NoEncoder,MinMaxScaler,PolynomialFeatures | no embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.22 |
| 1 | SimpleImputer,OneHotEncoder,NoScaler,KernelPCA | no embedding,ShapedResNetBackbone,FullyConnectedHead,nn.Sequential | 0.14 |
| 2 | None | CBLearner | 0.14 |
| 3 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential | 0.14 |
| 4 | SimpleImputer,OneHotEncoder,MinMaxScaler,NoFeaturePreprocessing | embedding,ShapedResNetBackbone,FullyConnectedHead,nn.Sequential | 0.08 |
| 5 | None | ETLearner | 0.06 |
| 6 | SimpleImputer,OneHotEncoder,StandardScaler,NoFeaturePreprocessing | no embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential | 0.06 |
| 7 | SimpleImputer,OneHotEncoder,MinMaxScaler,PowerTransformer | embedding,ShapedMLPBackbone,FullyConnectedHead,nn.Sequential | 0.04 |
| 8 | None | SVMLearner | 0.04 |
| 9 | None | KNNLearner | 0.04 |
| 8 | None | RFLearner | 0.04 |
| 9 | None | SVMLearner | 0.04 |
| 10 | None | KNNLearner | 0.02 |
| 11 | 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 22.454 seconds)
**Total running time of the script:** ( 5 minutes 30.492 seconds)


.. _sphx_glr_download_examples_20_basics_example_tabular_classification.py:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -125,7 +125,7 @@ Search for an ensemble of machine learning algorithms
.. code-block:: none
<autoPyTorch.api.tabular_regression.TabularRegressionTask object at 0x7fe106ad64f0>
<autoPyTorch.api.tabular_regression.TabularRegressionTask object at 0x7ff273a71a90>
Expand Down Expand Up @@ -157,7 +157,7 @@ Print the final ensemble performance

.. code-block:: none
<smac.runhistory.runhistory.RunHistory object at 0x7fe110cfc7f0> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
<smac.runhistory.runhistory.RunHistory object at 0x7ff273a71e80> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 64
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -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=0.0013153553009033203, budget=0), TrajEntry(train_perf=0.3195903909944855, incumbent_id=1, incumbent=Configuration:
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001180887222290039, 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'
Expand Down Expand Up @@ -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=2.4860830307006836, wallclock_time=3.518846035003662, budget=5.555555555555555)]
, ta_runs=1, ta_time_used=2.12186336517334, wallclock_time=3.1505396366119385, budget=5.555555555555555)]
{'r2': 0.943135905978891}
| | Preprocessing | Estimator | Weight |
|---:|:------------------------------------------------------------------|:----------------------------------------------------------------|---------:|
Expand All @@ -235,7 +235,7 @@ Print the final ensemble performance
.. rst-class:: sphx-glr-timing

**Total running time of the script:** ( 5 minutes 41.542 seconds)
**Total running time of the script:** ( 5 minutes 33.018 seconds)


.. _sphx_glr_download_examples_20_basics_example_tabular_regression.py:
Expand Down
Original file line number Diff line number Diff line change
Expand Up @@ -5,12 +5,12 @@

Computation times
=================
**11:10.158** total execution time for **examples_20_basics** files:
**11:09.365** total execution time for **examples_20_basics** files:

+--------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_examples_20_basics_example_tabular_regression.py` (``example_tabular_regression.py``) | 05:41.542 | 0.0 MB |
| :ref:`sphx_glr_examples_20_basics_example_tabular_regression.py` (``example_tabular_regression.py``) | 05:33.018 | 0.0 MB |
+--------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_examples_20_basics_example_tabular_classification.py` (``example_tabular_classification.py``) | 05:22.454 | 0.0 MB |
| :ref:`sphx_glr_examples_20_basics_example_tabular_classification.py` (``example_tabular_classification.py``) | 05:30.492 | 0.0 MB |
+--------------------------------------------------------------------------------------------------------------+-----------+--------+
| :ref:`sphx_glr_examples_20_basics_example_image_classification.py` (``example_image_classification.py``) | 00:06.162 | 0.0 MB |
| :ref:`sphx_glr_examples_20_basics_example_image_classification.py` (``example_image_classification.py``) | 00:05.855 | 0.0 MB |
+--------------------------------------------------------------------------------------------------------------+-----------+--------+
Original file line number Diff line number Diff line change
Expand Up @@ -46,7 +46,7 @@ the search. Currently, there are two changes that can be made to the space:-

.. code-block:: none
<smac.runhistory.runhistory.RunHistory object at 0x7fe06d866f10> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
<smac.runhistory.runhistory.RunHistory object at 0x7ff1cfc528e0> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 32
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -80,7 +80,7 @@ the search. Currently, there are two changes that can be made to the space:-
scaler:__choice__, Value: 'StandardScaler'
trainer:StandardTrainer:weighted_loss, Value: True
trainer:__choice__, Value: 'StandardTrainer'
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.0013649463653564453, budget=0), TrajEntry(train_perf=0.15204678362573099, incumbent_id=1, incumbent=Configuration:
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001199960708618164, budget=0), TrajEntry(train_perf=0.15204678362573099, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 32
encoder:__choice__, Value: 'OneHotEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -114,7 +114,7 @@ the search. Currently, there are two changes that can be made to the space:-
scaler:__choice__, Value: 'StandardScaler'
trainer:StandardTrainer:weighted_loss, Value: True
trainer:__choice__, Value: 'StandardTrainer'
, ta_runs=1, ta_time_used=4.133220195770264, wallclock_time=5.185335159301758, budget=5.555555555555555)]
, ta_runs=1, ta_time_used=3.7214386463165283, wallclock_time=4.764085292816162, budget=5.555555555555555)]
{'accuracy': 0.8670520231213873}
| | Preprocessing | Estimator | Weight |
|---:|:------------------------------------------------------------------|:----------------------------------------------------------|---------:|
Expand All @@ -128,7 +128,7 @@ the search. Currently, there are two changes that can be made to the space:-
| 7 | None | ETLearner | 0.08 |
| 8 | None | SVMLearner | 0.06 |
| 9 | SimpleImputer,OneHotEncoder,StandardScaler,KitchenSink | embedding,MLPBackbone,FullyConnectedHead,nn.Sequential | 0.04 |
<smac.runhistory.runhistory.RunHistory object at 0x7fe06d995f10> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
<smac.runhistory.runhistory.RunHistory object at 0x7ff1d0475070> [TrajEntry(train_perf=2147483648, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 32
encoder:__choice__, Value: 'NoEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -160,7 +160,7 @@ the search. Currently, there are two changes that can be made to the space:-
scaler:__choice__, Value: 'StandardScaler'
trainer:StandardTrainer:weighted_loss, Value: True
trainer:__choice__, Value: 'StandardTrainer'
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.0012934207916259766, budget=0), TrajEntry(train_perf=0.16959064327485385, incumbent_id=1, incumbent=Configuration:
, ta_runs=0, ta_time_used=0.0, wallclock_time=0.001115560531616211, budget=0), TrajEntry(train_perf=0.16959064327485385, incumbent_id=1, incumbent=Configuration:
data_loader:batch_size, Value: 32
encoder:__choice__, Value: 'NoEncoder'
feature_preprocessor:__choice__, Value: 'NoFeaturePreprocessor'
Expand Down Expand Up @@ -192,7 +192,7 @@ the search. Currently, there are two changes that can be made to the space:-
scaler:__choice__, Value: 'StandardScaler'
trainer:StandardTrainer:weighted_loss, Value: True
trainer:__choice__, Value: 'StandardTrainer'
, ta_runs=1, ta_time_used=2.6655709743499756, wallclock_time=3.714816093444824, budget=5.555555555555555)]
, ta_runs=1, ta_time_used=2.4923224449157715, wallclock_time=3.534742832183838, budget=5.555555555555555)]
{'accuracy': 0.861271676300578}
| | Preprocessing | Estimator | Weight |
|---:|:--------------------------------------------------------------|:-------------------------------------------------------------------|---------:|
Expand Down Expand Up @@ -337,7 +337,7 @@ the search. Currently, there are two changes that can be made to the space:-
.. rst-class:: sphx-glr-timing

**Total running time of the script:** ( 5 minutes 43.500 seconds)
**Total running time of the script:** ( 5 minutes 36.129 seconds)


.. _sphx_glr_download_examples_40_advanced_example_custom_configuration_space.py:
Expand Down

0 comments on commit 8f70383

Please sign in to comment.