..
adadelta_op_gpu.cu
adadelta_op.cc
adadelta_op.h
adagrad_fused_op_gpu.cu
[caffe2] Optimize Dedup version of RowWiseSparseAdagrad fused op by WarpReduce ( #45649 )
2020-10-02 14:28:24 -07:00
adagrad_fused_op_gpu.cuh
[caffe2] Optimize Dedup version of RowWiseSparseAdagrad fused op by WarpReduce ( #45649 )
2020-10-02 14:28:24 -07:00
adagrad_fused.cc
SparseAdagrad/RowWiseSparseAdagrad mean fusion on CPU & GPU and dedup version for RowWiseSparse mean fusion on GPU
2020-07-14 22:36:16 -07:00
adagrad_fused.h
SparseAdagrad/RowWiseSparseAdagrad mean fusion on CPU & GPU and dedup version for RowWiseSparse mean fusion on GPU
2020-07-14 22:36:16 -07:00
adagrad_op_gpu.cu
Fix the flooding log issues ( #38356 )
2020-05-13 11:23:17 -07:00
adagrad_op.cc
Fix potential divide by zero for CostInferenceForRowWiseSparseAdagrad
2020-09-04 14:14:49 -07:00
adagrad_op.h
[Caffe2] use the real new fbgemm sparse adagrad interface ( #46132 )
2020-10-10 08:57:54 -07:00
adam_op_gpu.cu
Support RowWiseSparseAdam on GPU ( #35404 )
2020-07-31 10:47:29 -07:00
adam_op.cc
Support RowWiseSparseAdam on GPU ( #35404 )
2020-07-31 10:47:29 -07:00
adam_op.h
[caffe2] compute r_correction only for radam to avoid sqrt(negative) ( #39393 )
2020-06-03 19:09:28 -07:00
clip_tensor_op.cc
Fix typos, via a Levenshtein-type corrector ( #31523 )
2020-01-17 16:03:19 -08:00
clip_tensor_op.h
CMakeLists.txt
fp16_momentum_sgd_op.cu
fp16_momentum_sgd_op.h
fp32_momentum_sgd_op.cu
fp32_momentum_sgd_op.h
ftrl_op.cc
ftrl_op.h
gftrl_op.cc
gftrl_op.h
iter_op_gpu.cc
iter_op.cc
iter_op.h
Fix the flooding log issues ( #38356 )
2020-05-13 11:23:17 -07:00
lars_op_gpu.cu
lars_op.cc
lars_op.h
learning_rate_adaption_op.cc
learning_rate_adaption_op.h
learning_rate_functors.h
[GradualGating] support better k value change ( #41557 )
2020-07-17 20:44:28 -07:00
learning_rate_op_gpu.cc
learning_rate_op.cc
[dper3] Create dper LearningRate low-level module ( #44639 )
2020-09-22 08:26:07 -07:00
learning_rate_op.h
[c2] Fix int types for learning rate
2020-09-26 10:59:29 -07:00
math_lp.cc
[caffe2] move fused SparseAdagrad to open source ( #35164 )
2020-03-26 17:29:12 -07:00
math_lp.h
[caffe2] move fused SparseAdagrad to open source ( #35164 )
2020-03-26 17:29:12 -07:00
momentum_sgd_op_gpu.cu
momentum_sgd_op.cc
momentum_sgd_op.h
Expose the interface of nesterov of SGD Optimizer from caffe2 to dper
2020-09-09 19:37:00 -07:00
rmsprop_op_gpu.cu
rmsprop_op.cc
rmsprop_op.h
rowwise_adagrad_fused.cc
SparseAdagrad/RowWiseSparseAdagrad mean fusion on CPU & GPU and dedup version for RowWiseSparse mean fusion on GPU
2020-07-14 22:36:16 -07:00
rowwise_adagrad_fused.h
[caffe2] Fix preprocessor checks for FMA
2020-10-11 11:48:32 -07:00
rowwise_counter.cc
Enforce counter value to double type in rowwise_counter
2020-08-05 20:40:51 -07:00
rowwise_counter.h
Enforce counter value to double type in rowwise_counter
2020-08-05 20:40:51 -07:00
storm_op.cc
Implementation and operator test for STORM optimizer ( #36225 )
2020-04-14 23:04:26 -07:00
storm_op.h
Implementation and operator test for STORM optimizer ( #36225 )
2020-04-14 23:04:26 -07:00
weight_scale_op_gpu.cc
[c2] cuda impl for WeightScale op ( #38712 )
2020-05-26 12:50:54 -07:00
weight_scale_op.cc
[c2] cuda impl for WeightScale op ( #38712 )
2020-05-26 12:50:54 -07:00
weight_scale_op.h
[c2] cuda impl for WeightScale op ( #38712 )
2020-05-26 12:50:54 -07:00
wngrad_op.cc
wngrad_op.h
yellowfin_op_gpu.cu
yellowfin_op.cc
yellowfin_op.h
1a99689d71