optimizer exploration - v1 and v2 + fix position_weighted optimizer + decoupled weight decay #53881
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
… decoupled weight decay Summary: 1. Fix position_weighted optimizer: Position weighted layer uses default optimizer but is actually gradient_slice, which will cause problem if we do not handle it properly in the new optimizier. The solution is to use sparseadagrad when it is gradient_slices. 2. Optimizer implementation of v1 and v2: using 1st momentum with/without bias_correction. 3. also implemented decoupled weight decay in the new optimizer. Test Plan: buck test //caffe2/caffe2/fb/dper/layer_models/tests/split_1:sparse_nn_test_2 -- test_mlp_optimization buck test //caffe2/caffe2/python:optimizer_test -- TestDecayAdagrad buck test //caffe2/caffe2/python/operator_test:decay_adagrad_test ctr_mbl_feed work flow: f255731660 oc work flow: f255739503 Reviewed By: 0x10cxR1 Differential Revision: D26839668 fbshipit-source-id: 3e0a3646d8459c769caea19658217f1a32d539bb
💊 CI failures summary and remediationsAs of commit 929ae4b (more details on the Dr. CI page):
ci.pytorch.org: 1 failedThis comment was automatically generated by Dr. CI (expand for details).Follow this link to opt-out of these comments for your Pull Requests.Please report bugs/suggestions to the (internal) Dr. CI Users group. |
|
This pull request was exported from Phabricator. Differential Revision: D26839668 |
Codecov Report
@@ Coverage Diff @@
## master #53881 +/- ##
==========================================
- Coverage 77.30% 77.30% -0.01%
==========================================
Files 1888 1888
Lines 183589 183589
==========================================
- Hits 141923 141918 -5
- Misses 41666 41671 +5 |
|
This pull request was exported from Phabricator. Differential Revision: D26839668 |
lanlanfb
added a commit
to lanlanfb/pytorch
that referenced
this pull request
Mar 25, 2021
… decoupled weight decay (pytorch#54042) Summary: Pull Request resolved: pytorch#54042 Pull Request resolved: pytorch#53881 1. Fix position_weighted optimizer: Position weighted layer uses default optimizer but is actually gradient_slice, which will cause problem if we do not handle it properly in the new optimizier. The solution is to use sparseadagrad when it is gradient_slices. 2. Optimizer implementation of v1 and v2: using 1st momentum with/without bias_correction. 3. also implemented decoupled weight decay in the new optimizer. Test Plan: buck test //caffe2/caffe2/fb/dper/layer_models/tests/split_1:sparse_nn_test_2 -- test_mlp_optimization buck test //caffe2/caffe2/python:optimizer_test -- TestDecayAdagrad buck test //caffe2/caffe2/python/operator_test:decay_adagrad_test ctr_mbl_feed work flow: f255731660 oc work flow: f255739503 Reviewed By: 0x10cxR1 Differential Revision: D26839668 fbshipit-source-id: 8a2170e317e695b861b1b1e566beb82ae0f08836
facebook-github-bot
pushed a commit
that referenced
this pull request
Mar 28, 2021
… decoupled weight decay (#54042) Summary: Pull Request resolved: #54042 Pull Request resolved: #53881 1. Fix position_weighted optimizer: Position weighted layer uses default optimizer but is actually gradient_slice, which will cause problem if we do not handle it properly in the new optimizier. The solution is to use sparseadagrad when it is gradient_slices. 2. Optimizer implementation of v1 and v2: using 1st momentum with/without bias_correction. 3. also implemented decoupled weight decay in the new optimizer. Test Plan: buck test //caffe2/caffe2/fb/dper/layer_models/tests/split_1:sparse_nn_test_2 -- test_mlp_optimization buck test //caffe2/caffe2/python:optimizer_test -- TestDecayAdagrad buck test //caffe2/caffe2/python/operator_test:decay_adagrad_test ctr_mbl_feed work flow: f255731660 oc work flow: f255739503 Reviewed By: 0x10cxR1 Differential Revision: D26839668 fbshipit-source-id: 2b6881c1a88540ef5766be40f5e80001257e2199
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary:
Test Plan:
buck test //caffe2/caffe2/fb/dper/layer_models/tests/split_1:sparse_nn_test_2 -- test_mlp_optimization
buck test //caffe2/caffe2/python:optimizer_test -- TestDecayAdagrad
buck test //caffe2/caffe2/python/operator_test:decay_adagrad_test
ctr_mbl_feed work flow: f255731660
oc work flow: f255739503
Reviewed By: 0x10cxR1
Differential Revision: D26839668