Paddle

History

littletomatodonkey 1c08a2136e test=develop, add addmm op (#23384 ) add addmm op		5 years ago
..
detail	Add dependency for error_codes.proto (#21084 )	5 years ago
CMakeLists.txt	[Paddle-TRT]: Ernie Dynamic shape support. (#23138 )	5 years ago
algorithm.h	…
beam_search.cc	Make parent_idx a dispensable output for beam_search op to support models saved by older paddle version. (#16106 )	6 years ago
beam_search.cu	Make parent_idx a dispensable output for beam_search op to support models saved by older paddle version. (#16106 )	6 years ago
beam_search.h	Return parent_idx in beam_search op (#15520 )	6 years ago
beam_search_test.cc	Return parent_idx in beam_search op (#15520 )	6 years ago
bert_encoder_functor.cu	[Paddle-TRT]: Ernie Dynamic shape support. (#23138 )	5 years ago
bert_encoder_functor.h	[Paddle-TRT]: Ernie Dynamic shape support. (#23138 )	5 years ago
blas.cc	…
blas.h	Improve elementwise operators performance in same dimensions. (#19763 )	5 years ago
blas_impl.cu.h	test=develop, add addmm op (#23384 )	5 years ago
blas_impl.h	fix typo word (#22784 )	5 years ago
bloomfilter.h	refine murmurhash3_x64_128 for bloom_filter (#20996 )	5 years ago
compound_functors.h	Optimize fused_elewise_activation_grad op. (#18041 )	6 years ago
concat.hip.cu	…
concat_and_split.cc	…
concat_and_split.cu	Replace TemporaryAllocator by CUDADeviceContextAllocator (#18989 )	6 years ago
concat_and_split.h	…
concat_test.cc	Optimize the concat and split kernel for specical cases when the number of inputs/outputs is 2 (#17415 )	6 years ago
context_project.cc	…
context_project.cu	Fix Eigen macro when using GPU	6 years ago
context_project.h	fix security bugs : (#17464 )	6 years ago
cos_sim_functor.cc	…
cos_sim_functor.cu	…
cos_sim_functor.h	…
cpu_vec.h	fix sofmax seg fault in AVX, test=develop (#19487 )	6 years ago
cpu_vec_test.cc	fix -Wno-error=sign-compare warning in gcc8 (#21434 )	5 years ago
cross_entropy.cc	polish cross_entropy ENFORCE (#22056 )	5 years ago
cross_entropy.cu	unify PADDLE_ASSERT_MSG into PADDLE_ENFORCE(error_message) (#19631 )	6 years ago
cross_entropy.h	unify PADDLE_ASSERT_MSG into PADDLE_ENFORCE(error_message) (#19631 )	6 years ago
depthwise_conv.cu	Optimize the depthwise op test=develop (#22265 )	5 years ago
depthwise_conv.h	conv_transpose supports channel_last input, test=develop, test=document_preview (#20072 )	5 years ago
fc.cc	optimize fc jit (#21878 )	5 years ago
fc.cu	[Ernie GPU Optimize]: Embedding_eltwise_layernorm Fuse (#22494 )	5 years ago
fc.h	Add fc padding to improve mkl GEMM's performance when N and K are multiple of 128. (#20972 )	5 years ago
functors.h	Support fp16 in GPU impl of fused_elemwise_activation_op. (#20636 )	5 years ago
gru_compute.cc	fix build problem test=develop	6 years ago
gru_compute.cu	Fix ce ocr_recognition test fails (#20987 )	5 years ago
gru_compute.h	update gru op forward kernel	6 years ago
im2col.cc	fix the bug of conv_transpose:compatible with Anylayout setting, test=develop (#20897 )	5 years ago
im2col.cu	fix the bug of conv_transpose:compatible with Anylayout setting, test=develop (#20897 )	5 years ago
im2col.h	conv_transpose supports channel_last input, test=develop, test=document_preview (#20072 )	5 years ago
im2col_cfo_cpu.h	fix conv_transpose's bug: compatible with Anylayout setting, test=develop (#20589 )	5 years ago
im2col_test.cc	…
lstm_compute.cc	add cell clip and proj clip, fix bug for h0	6 years ago
lstm_compute.cu	add cell clip and proj clip, fix bug for h0	6 years ago
lstm_compute.h	add cell clip and proj clip, fix bug for h0	6 years ago
math_cuda_utils.h	[Paddle-TRT]: Ernie Dynamic shape support. (#23138 )	5 years ago
math_function.cc	Change tensor uses proto::VarType::type	6 years ago
math_function.cu	Sequence mask support tensor (#18249 )	6 years ago
math_function.h	…
math_function_impl.h	rewrite ddim	6 years ago
math_function_test.cc	…
math_function_test.cu	…
matrix_bit_code.cc	fix bug after merge reyoung optimization, test=develop	6 years ago
matrix_bit_code.h	fix typo word (#22784 )	5 years ago
maxouting.cc	maxout supports channel_last input (#20846 )	5 years ago
maxouting.cu	maxout supports channel_last input (#20846 )	5 years ago
maxouting.h	maxout supports channel_last input (#20846 )	5 years ago
padding.h	remove -Wno-error=sign-compare, make warning as error (#21358 )	5 years ago
pooling.cc	fix pool2d pool3d,support asymmetric padding and channel_last (#19739 )	5 years ago
pooling.cu	Bug Fix: Paddle-TRT cannot handle adaptive pooling in pool2d op converter and "num" attribute in split op converter (#20733 )	5 years ago
pooling.h	Bug Fix: Paddle-TRT cannot handle adaptive pooling in pool2d op converter and "num" attribute in split op converter (#20733 )	5 years ago
prelu.cu	fix the computation for dx (grad for x) for prelu operation. (#20949 )	5 years ago
prelu.h	fix the computation for dx (grad for x) for prelu operation. (#20949 )	5 years ago
sample_prob.cc	update comment	6 years ago
sample_prob.cu	refine PADDLE_ENFORCE codes for unify PADDLE_ASSERT_MSG (#19603 )	6 years ago
sample_prob.h	update comment	6 years ago
sampler.cc	nce add check sample lables, test=develop (#15463 )	6 years ago
sampler.h	nce add check sample lables, test=develop (#15463 )	6 years ago
selected_rows_functor.cc	fix select_rows mergeadd bug, test=develop (#20876 )	5 years ago
selected_rows_functor.cu	refine PADDLE_ENFORCE codes for unify PADDLE_ASSERT_MSG (#19603 )	6 years ago
selected_rows_functor.h	fix the diff between async mode and async_half mode (#19535 )	6 years ago
selected_rows_functor_test.cc	fix the diff between async mode and async_half mode (#19535 )	6 years ago
selected_rows_functor_test.cu.cc	Add the CUDA kernel for beam_search op (#15020 )	6 years ago
sequence2batch.cc	…
sequence2batch.cu	Fix Eigen macro when using GPU	6 years ago
sequence2batch.h	Fix the integer overflow problem of sequence2batch (#22479 )	5 years ago
sequence_padding.cc	Optimize the sequence padding op (#17403 )	6 years ago
sequence_padding.cu	fix sequence pad; test=develop	6 years ago
sequence_padding.h	Fix warpctc in padding mode. (#21033 )	5 years ago
sequence_padding_test.cc	Correct the use of DeviceContext in unittest sequence_pooling_test and sequence_padding_test (#22456 )	5 years ago
sequence_pooling.cc	support 2-level lod of input in sequence_pool (#19839 )	5 years ago
sequence_pooling.cu	Fix dismatch of std::max's arguments type on windows. (#22507 )	5 years ago
sequence_pooling.h	support 2-level lod of input in sequence_pool (#19839 )	5 years ago
sequence_pooling_test.cc	Correct the use of DeviceContext in unittest sequence_pooling_test and sequence_padding_test (#22456 )	5 years ago
sequence_scale.cc	…
sequence_scale.cu	…
sequence_scale.h	…
softmax.cc	…
softmax.cu	replace CUDNN_ENFORCE with PADDLE_ENFORCE_CUDA_SUCCESS, test=develop (#22109 )	5 years ago
softmax.h	[CPU] refine cpu softmax bwd (#17534 )	6 years ago
softmax_impl.h	fix softmax CE time limit check failed (#19846 )	5 years ago
tree2col.cc	Tree conv op (#15217 )	6 years ago
tree2col.cu	Tree conv op (#15217 )	6 years ago
tree2col.h	Tree conv op (#15217 )	6 years ago
unpooling.cc	…
unpooling.cu	unify PADDLE_ASSERT_MSG into PADDLE_ENFORCE(error_message) (#19631 )	6 years ago
unpooling.h	…
vol2col.cc	fix conv_transpose's bug: compatible with Anylayout setting, test=develop (#20589 )	5 years ago
vol2col.cu	fix conv_transpose's bug: compatible with Anylayout setting, test=develop (#20589 )	5 years ago
vol2col.h	conv_transpose supports channel_last input, test=develop, test=document_preview (#20072 )	5 years ago
vol2col_test.cc	…