Merge pull request #275 from microsoft/master

merge master
SparkSnail · Nov 16, 2020 · f73367f · f73367f
2 parents 7eb15f8 + fcbf05b
commit f73367f
Show file tree

Hide file tree

Showing 214 changed files with 1,429 additions and 987 deletions.
diff --git a/.gitignore b/.gitignore
@@ -5,6 +5,10 @@
 /test/model_path/
 /test/temp.json
 /test/ut/sdk/*.pth
+/test/ut/tools/annotation/_generated/
+/ts/nni_manager/exp_profile.json
+/ts/nni_manager/metrics.json
+/ts/nni_manager/trial_jobs.json
 
 
 # Logs

diff --git a/archive-ut/nni_annotation/test_annotation.py b/archive-ut/nni_annotation/test_annotation.py
diff --git a/archive-ut/nni_cmd/tests/mock/nnictl_metadata/.experiment b/archive-ut/nni_cmd/tests/mock/nnictl_metadata/.experiment
diff --git a/archive-ut/nni_cmd/tests/mock/nnictl_metadata/config/.config b/archive-ut/nni_cmd/tests/mock/nnictl_metadata/config/.config
diff --git a/archive-ut/nni_cmd/tests/test_common_utils.py b/archive-ut/nni_cmd/tests/test_common_utils.py
diff --git a/archive-ut/nni_trial_tool/test/test_file_channel.py b/archive-ut/nni_trial_tool/test/test_file_channel.py
diff --git a/archive-ut/nni_trial_tool/test/test_hdfsClientUtility.py b/archive-ut/nni_trial_tool/test/test_hdfsClientUtility.py
diff --git a/docs/en_US/Compression/AutoPruningUsingTuners.md b/docs/en_US/Compression/AutoPruningUsingTuners.md
@@ -7,13 +7,13 @@ It's convenient to implement auto model pruning with NNI compression and NNI tun
 You can easily compress a model with NNI compression. Take pruning for example, you can prune a pretrained model with LevelPruner like this
 
 ```python
-from nni.compression.torch import LevelPruner
+from nni.algorithms.compression.pytorch.pruning import LevelPruner
 config_list = [{ 'sparsity': 0.8, 'op_types': ['default'] }]
 pruner = LevelPruner(model, config_list)
 pruner.compress()
 ```
 
-The 'default' op_type stands for the module types defined in [default_layers.py](https://github.com/microsoft/nni/blob/v1.9/src/sdk/pynni/nni/compression/torch/default_layers.py) for pytorch.
+The 'default' op_type stands for the module types defined in [default_layers.py](https://github.com/microsoft/nni/blob/v1.9/src/sdk/pynni/nni/compression/pytorch/default_layers.py) for pytorch.
 
 Therefore ```{ 'sparsity': 0.8, 'op_types': ['default'] }```means that **all layers with specified op_types will be compressed with the same 0.8 sparsity**. When ```pruner.compress()``` called, the model is compressed with masks and after that you can normally fine tune this model and **pruned weights won't be updated** which have been masked.
 
@@ -71,7 +71,7 @@ Then we need to modify our codes for few lines
 
 ```python
 import nni
-from nni.compression.torch import *
+from nni.algorithms.compression.pytorch.pruning import *
 params = nni.get_parameters()
 conv0_sparsity = params['prune_method']['conv0_sparsity']
 conv1_sparsity = params['prune_method']['conv1_sparsity']

diff --git a/docs/en_US/Compression/CompressionReference.md b/docs/en_US/Compression/CompressionReference.md
@@ -7,34 +7,34 @@
 ## Sensitivity Utilities
 
 ```eval_rst
-..  autoclass:: nni.compression.torch.utils.sensitivity_analysis.SensitivityAnalysis
+..  autoclass:: nni.compression.pytorch.utils.sensitivity_analysis.SensitivityAnalysis
     :members:
 
 ```
 
 ## Topology Utilities
 
 ```eval_rst
-..  autoclass:: nni.compression.torch.utils.shape_dependency.ChannelDependency
+..  autoclass:: nni.compression.pytorch.utils.shape_dependency.ChannelDependency
     :members:
 
-..  autoclass:: nni.compression.torch.utils.shape_dependency.GroupDependency
+..  autoclass:: nni.compression.pytorch.utils.shape_dependency.GroupDependency
     :members:
 
-..  autoclass:: nni.compression.torch.utils.mask_conflict.CatMaskPadding
+..  autoclass:: nni.compression.pytorch.utils.mask_conflict.CatMaskPadding
     :members:
 
-..  autoclass:: nni.compression.torch.utils.mask_conflict.GroupMaskConflict
+..  autoclass:: nni.compression.pytorch.utils.mask_conflict.GroupMaskConflict
     :members:
 
-..  autoclass:: nni.compression.torch.utils.mask_conflict.ChannelMaskConflict
+..  autoclass:: nni.compression.pytorch.utils.mask_conflict.ChannelMaskConflict
     :members:
 
 ```
 
 ## Model FLOPs/Parameters Counter
 
 ```eval_rst
-..  autofunction:: nni.compression.torch.utils.counter.count_flops_params
+..  autofunction:: nni.compression.pytorch.utils.counter.count_flops_params
 
-```
+```