COMMITS
November 29, 2019
H
squad multi-lamb
Haibin Lin committed
November 2, 2019
R
multi-precision lamb update operator
Rohit Kumar Srivastava committed
November 12, 2019
D
Speed fused_op compilation by caching ptx and jit-compiled functions (#16783)
Dick Carter committed
November 15, 2019
R
adding API doc for Lamb Phase 1 and 2
Rohit Kumar Srivastava committed
November 13, 2019
R
fixing base lamb optimizer
Rohit Kumar Srivastava committed
October 29, 2019
A
initial commit lamb optimizer
Anirudh Acharya committed
October 2, 2019
C
fix atol for test_preloaded_multi_sgd (#16356)
Caenorst committed
September 30, 2019
C
Add fast implementation of LARS (#16122)
Caenorst committed
October 22, 2019
S
Make mrcnn_mask_target arg mask_size a 2d tuple (#16567)
Serge Panev committed
October 17, 2019
L
Load NDArray only to GPU if GPU is present (#16432)
Leonard Lausen committed
October 15, 2019
S
Move MRCNNMaskTarget op to contrib (#16486)
Serge Panev committed
C
add better error message and description and remove cutlass from compilation flags
cfujitsang committed
C
remove all cutlass stuff
cfujitsang committed
October 24, 2019
P
Fix for wrong reqs set after switching from training to inference (#16553)
Przemyslaw Tredak committed
November 1, 2019
C
fix another typo
cfujitsang committed
C
remove bwd_ignore_zero_init in test_operator_gpu.py
cfujitsang committed
C
remove bwd_ignore_zero_init flag
cfujitsang committed
C
remove std::move for clang
cfujitsang committed
C
add error return if bwd_ignore_zero_init is used without MXNET_EXEC_ENABLE_ADDTO
cfujitsang committed
October 31, 2019
C
add warning in bwd_ignore_zero_init description, also test with fp32
cfujitsang committed
October 21, 2019
C
add _contrib for namespace and add GPU only on description
cfujitsang committed
C
remove a wrong copy-paste
cfujitsang committed
C
add equivalent mxnet code in description of mha ops
cfujitsang committed
October 18, 2019
C
only use cublasStridedBatchedGemm if CUDA >= 9.1
cfujitsang committed
C
remove tests using kAddTo
cfujitsang committed
C
Add all the cuda/cublas helper functions
cfujitsang committed
October 16, 2019
C
correct another typo
cfujitsang committed
C
change credit for the approach since the code have changed
cfujitsang committed