- 30 Apr, 2020 1 commit
-
-
Deyu Fu authored
* modify MTA axpby for wider load/store * Make scale/axpby/l2/adam/lamb multi_tensor uses wider load
-
- 26 Apr, 2019 1 commit
-
-
ptrblck authored
* change .type().ScalarType() to .scalar_type() + at::ScalarType::X to at::kX * revert scalar_type() to type() for AT_DISPATCH_FLOATING_TYPES_AND_HALF * revert scalar_type() to type() in AT_DISPATCH_FLOATING_TYPES * revert scalar_type() to type() for AT_DISPATCH_FLOATING_TYPES_AND_HALF in welford.cu * revert scalar_type() to type() in layer_norm_cuda_kernel.cu * revert at::kType to at::ScalarType::Type * use DISPATCH_FLOAT_AND_HALF to get rid of warnings * add dispatch mechanisms for double+float and double+float+half
-
- 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
-
- 15 Mar, 2019 1 commit
-
-
Michael Carilli authored
-
- 12 Mar, 2019 1 commit
-
-
Michael Carilli authored
-
- 10 Mar, 2019 1 commit
-
-
Natalia Gimelshein authored
-
- 22 Feb, 2019 1 commit
-
-
Michael Carilli authored
Allow multi-tensor unscale to handle FP16 output, so it can also be used for copy-scatter. Rename some options.
-
- 19 Feb, 2019 1 commit
-
-
Michael Carilli authored
-