Commit Graph

58 Commits (a97d5a6153ac972435e6f2e6915c180400742fe7)

Author SHA1 Message Date
Chen Weihang aa0f254fbe
Add macro BOOST_GET to enrich the error information of boost :: get (#24175)
5 years ago
0YuanZhang0 a164b10d05
API/OP error message enhancement (#23717)
5 years ago
Chen Weihang 16315d3d9e
Delete Ref & VectorRef and add GetDataSafely (#22997)
5 years ago
Zeng Jinle b8886bf122
rename no_need_buffer_vars_macro, test=develop (#23159)
5 years ago
Zeng Jinle d33c4343e1
Imperative tracer refactoring (#22457)
5 years ago
tianshuo78520a 433cef03e5
fix typo word (#22784)
5 years ago
tianshuo78520a d2ba91aad1
fix typo words (#22653)
5 years ago
huzhiqiang 9e29d3ebed
【OpPorting Example】DEMO OF FIX COMPILE&RUNTIME LOD_EQUALITY (#22460)
5 years ago
Aurelius84 5cb2c74127
add register op_data_type of pad/expand_as et.al (#21718)
5 years ago
Zeng Jinle 0f8888360e
Polish op registry codes (#21561)
5 years ago
Aurelius84 3d9dee575e
Set lod_level of Out in compile time of sequence_pool_op (#21604)
5 years ago
Chen Weihang 7e5f74b825
Add examples for error message writing specification - InvalidArgument (#21132)
5 years ago
Yiqun Liu 35f17ae28f
Add the check of lod_level between compile-time and runtime. (#20961)
5 years ago
Zeng Jinle 878a40f57d
Support NoNeedBufferVarsInference in dygraph backward (#20868)
5 years ago
hong 8c4573a3cb
GradMaker for dygraph (#19706)
5 years ago
Yiqun Liu 6fcfd32e6c
Check and correct the output's lod_level in DynamicRNN related operators (#19144)
5 years ago
Chen Weihang 26cc1fe508
Replace risky GetInputType method with secure IndicateVarDataType interface (#20668)
5 years ago
Zeng Jinle 3eebd5b391
refine sequence_softmax grad maker, test=develop (#20127)
5 years ago
Li Fuchen 5365cd2f14
Set lod level of sequence_unpad's output to 1 in compile time (#20068)
5 years ago
Chen Weihang 1409586eaa Add LoD empty check for all related sequence ops (#19980)
5 years ago
Aurelius84 99a9615a4b
Removing length dims constraints of seq_pad and seq_unpad (#19497)
5 years ago
Aurelius84 fcf53e55ff
support 2-level lod of input in sequence_pool (#19839)
5 years ago
Tao Luo d6c85c96dc
paddle::framework::vectorize() templatization (#19627)
6 years ago
Aurelius84 5b5379b32a
Add sequence_topk_avg_pooling Op (#19442)
6 years ago
Yiqun Liu a445c33552
Add the check of lod in sequence_softmax kernel. (#18996)
6 years ago
Hongyu Liu df2eee71d8
Sequence mask support tensor (#18249)
6 years ago
Yibing Liu f57ee3693b
Fix the bug of sequence_unpad op (#18290)
6 years ago
wawltor 2ae8decc90
test=develop (#17984)
6 years ago
Yibing Liu 33d1e56506
Enable seq_pool op to accept len 0 input (#17284)
6 years ago
Guo Sheng 430e25654b
Fix the usage of out_grad lod in sequence_slice_op. (#17625)
6 years ago
Yihua Xu 218d8d8f73 Optimize the computing kernel of sequence_reverse operator (#17349)
6 years ago
Yibing Liu 3c375751f8
Support seq len equal to 0 in sequence ops (#16935)
6 years ago
Kevin c474e7ddf5 fix overflow by int32 mul test=develop (#16794)
6 years ago
Zeng Jinle d658244997
fix some grad op desc maker (#16581)
6 years ago
tensor-tang ead558b7f6
Merge pull request #16256 from tensor-tang/refine/seqenum
6 years ago
tensor-tang 50931dee1d refine seq enum op
6 years ago
luotao1 d9f0e7252a refine with comments
6 years ago
luotao1 721c2c00ef refine fc_infershape
6 years ago
sneaxiy 5a92e4c097 revert revert 16144
6 years ago
Zeng Jinle a91964c8fe Revert "PaddingRNN model memory optimize"
6 years ago
Zeng Jinle 0b49e43d3a
Merge pull request #16144 from sneaxiy/rnn_mem_opt
6 years ago
luotao1 fe78a92e6e refine with comments
6 years ago
sneaxiy b26e9bd232 refine code
6 years ago
luotao1 31ccaf0916 add all_kernels_must_compute_runtime_shape example for speedup infershape
6 years ago
whs 0f99d24083
Make sequence_erase op support for input with multi-level LoD. (#15982)
6 years ago
luotao1 34404f9c31 refine infershape of sequence_enumerate, hash and fuse_emb_seq_pool
6 years ago
Yiqun Liu f4634d76d7
Optimize the CUDA implementation of sequence_expand op by reduce the times of copying lod data from CPU to GPU. (#15493)
6 years ago
tensor-tang e1c707fe9c
fix warnings (#15790)
6 years ago
Yiqun Liu eaad3e4c3d
Add check of input in sequence_expand op. (#15466)
6 years ago
sneaxiy b56aca82e9 merge develop
6 years ago