Skip to content
GitLab
Menu
Projects
Groups
Snippets
Loading...
Help
Help
Support
Community forum
Keyboard shortcuts
?
Submit feedback
Contribute to GitLab
Sign in / Register
Toggle navigation
Menu
Open sidebar
mirrors
ROCmSoftwarePlatform
apex
Repository
Branches
Overview
Active
Stale
All
mlp_rocblas_alt_impl
c6aad5dd
·
Refactor rocblas_alt_impl implementation and only use it for backprop
·
Mar 23, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
mha_rocblas_alt_impl
e6df1c08
·
Fix for rocblas_alt_impl flag in MHA
·
Mar 24, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
spumma/fused-larc
58d89574
·
Update LARCFunctor's metadata argument type
·
Apr 05, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/rocblas_backward_compatible
f884c100
·
Fix some bugs
·
Apr 06, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
IFU-master-2022-02-09
eafa12c4
·
Fix compiling errors
·
Apr 09, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
apex_transformer
14ff43ad
·
Resolve filename collision of *cpp files with to-hipify code and *cu files
·
Apr 13, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/mha_faster_build
68c83071
·
Fix some bugs
·
Apr 13, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
IFU-master-2022-04-12
c3a9b360
·
Merge remote-tracking branch 'upstream/master' into IFU-master-2022-04-12
·
Apr 13, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/gakadam/apex-groupbn
cc38a72b
·
Added helper fns for ELEMENTS_PER_LDG=8.
·
Apr 22, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
ROCmBackwardPassGuard
695739bf
·
Use BACKWARD_PASS_GUARD_CLASS to prevent lengthy if-statement
·
Jun 01, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/fused_adam_debug
d6277b1d
·
TestFusedAdam fp16 debugging
·
Aug 02, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/FusedRMSNorm
fc79ed89
·
Use at::cuda::warp_size() instead of at::cuda::getCurrentDeviceProperties()->warpSize
·
Aug 03, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/skipped_tests
54937039
·
Update test_lamb.py
·
Aug 09, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/flaky_tests
merged
cebbb04f
·
Remove some comments in run_test.py
·
Aug 10, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
IFU-master-2022-07-29
merged
cc5f83b5
·
Skip a failing test introduced by a upstream PyTorch regression
·
Aug 11, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/fast_layer_norm
f2405243
·
Resolve some compiling issues on ROCm
·
Aug 18, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/cached_cast_fix
662ac624
·
Unskip the unit tests related to len(cached_x.grad_fn.next_functions) == 1
·
Aug 25, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/transducer
d8115964
·
Keep transducer extension CUDA-compatible
·
Sep 08, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/apex_peer_memory_nccl_p2p
merged
bc64ee83
·
Keep --peer_memory and --nccl_p2p CUDA-compatible
·
Sep 08, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
dev/hubertlu/focal_loss_and_index_mul_2d_cuda
merged
7a344314
·
Merge branch 'master' into dev/hubertlu/focal_loss_and_index_mul_2d_cuda
·
Sep 09, 2022
Compare
Select Archive Format
Download source code
zip
tar.gz
tar.bz2
tar
Prev
1
…
4
5
6
7
8
9
10
Next