- 13 Aug, 2019 2 commits
-
-
Deyu Fu authored
FusedSGD now work as before FusedAdam now work with o1/o2, no longer fuse scaling and casting Removed special backend handling for FusedAdam Moved and updated test for FusedAdam into run_optimizers Removed legacy tests for optimizers.FP16_optimizer and FusedAdam in run_mixed_adam
-
Marek Kolodziej authored
Co-authored-by:
Aditya Agrawal <aditya.iitb@gmail.com> Co-authored-by:
Marek Kolodziej <mkolod@gmail.com>
-
- 12 Aug, 2019 1 commit
-
-
Deyu Fu authored
-
- 08 Aug, 2019 1 commit
-
-
Deyu Fu authored
-
- 06 Aug, 2019 1 commit
-
-
ngimel authored
* Bug fix for non-affine layer-norm + add backward unit test * clean up tests and add tests for a large batch
-
- 03 Jul, 2019 1 commit
-
-
Michael Carilli authored
-
- 31 May, 2019 1 commit
-
-
mcarilli authored
* Existing tests passing, still need to add per-tensor tests * Test is passing, still need to measure performance * ILP for l2norm functor
-
- 27 May, 2019 2 commits
-
-
Michael Carilli authored
-
Michael Carilli authored
-
- 16 May, 2019 1 commit
-
-
mcarilli authored
* Support add_param_group * syntax * Test added and passing
-
- 02 May, 2019 1 commit
-
-
Michael Carilli authored
-
- 10 Apr, 2019 3 commits
-
-
Lam Dang authored
-
Lam Dang authored
-
Michael Carilli authored
-
- 04 Apr, 2019 1 commit
-
-
mcarilli authored
* Refactor to allow more flexible treatment of multiple optimizers/models/losses * Adding _process_optimizers.py * Created L0 tests (now passing). * fix: minor print typo (#234) * make L1 results easier to read * L0 multiple model/optimizer/loss test fleshed out * Adding test that master params remain synced across distributed processes * Docstring updates * Docstring updates
-
- 19 Mar, 2019 1 commit
-
-
Michael Carilli authored
-
- 13 Mar, 2019 1 commit
-
-
Wil Kong authored
-
- 10 Mar, 2019 1 commit
-
-
Michael Carilli authored
-
- 26 Feb, 2019 1 commit
-
-
Michael Carilli authored
-
