Commit Graph

6703 Commits (1041e18c4785159838c1531983432c507c291b2e)

Author SHA1 Message Date
tensor-tang 7044cfa7c7 add sgd jitcode and op test
6 years ago
tensor-tang 8e04133719 add benchmark and mkl sgd implement
6 years ago
tensor-tang 07efdb5139 Merge remote-tracking branch 'ups/develop' into jit/sgd
6 years ago
Jacek Czaja c63f6b2039 - MKL-DNN pooling updated to set_prim_desc
6 years ago
shippingwang 3398293272 add API.spec. test=develop
6 years ago
shippingwang 5ce46c637a fix api.spec, test=develop
6 years ago
qingqing01 8e439ccfff
Fix bug in fake_quantize_op and add more unit testing (#15912)
6 years ago
qingqing01 f4846bf3dc
loosly check in the InferShape of cross_entropy_op. (#15863)
6 years ago
minqiyang 28077c4da6 Add gperftools into imperative tracer
6 years ago
Yihua Xu 7396788694 Optimize gelu operation with mkl erf.
6 years ago
xiaoli.liu@intel.com 70759d181b Optimize INT8 DeQuantize Op with primitive reuse.
6 years ago
minqiyang efb2f2baf8 Fix bugs
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
Tao Luo 60546b78cc
Merge pull request #15923 from Sand3r-/mgallus/conv-residual-ut
6 years ago
guomingz 630c1e8317 This PR improve performance of prior_box op about 1.25x faster on CPU. (#15909)
6 years ago
Tao Luo 9c05421c97
Merge pull request #15914 from Sand3r-/mgallus/mkldnn-sum-code-reuse
6 years ago
chengduo 7ca8553d4e
Add alloc_continuous_space_op (#15900)
6 years ago
wopeizl 2192c46436
Merge pull request #15916 from wopeizl/win/fixevent1
6 years ago
baojun-nervana 2ffacdebc2 Update ngraph version to v0.14 test=develop
6 years ago
Michal Gallus 6a2bc9a275 Add Conv Residual Connection UT for Projection
6 years ago
Zhen Wang 548931456c update some functions' names according to the suggestion. test=develop
6 years ago
Michal Gallus 6ebe9877bb Improve code reuse at MKL-DNN sum
6 years ago
dzhwinter 660e410655
Merge pull request #15855 from dzhwinter/fix/nightly_test
6 years ago
peizhilin c6472579c0 test=develop
6 years ago
peizhilin b5d6e38b05 fix build issue for cudaEvent_t
6 years ago
minqiyang b420ec3a92 invoke backward_hooks after reduce op's depcounts map
6 years ago
Qiyang Min 4bd28b304b
Merge pull request #15831 from velconia/imperative_engine
6 years ago
Xin Pan a6e3cd5eb7
Merge pull request #15425 from panyx0718/api
6 years ago
liangan1 4acc522087 Enable function coverage for U8/S8 ConvMKLDNNOpKernel
6 years ago
wopeizl 3ccd8964a4
Merge pull request #15905 from wopeizl/win/fix_eigen
6 years ago
chengduo 8e904d322f
Remove unnecessary dependence for profiler (#15899)
6 years ago
Zhen Wang 9261cf39db update with develop. test=develop
6 years ago
Zhen Wang 0bf809c9b3 add set_attr for IrOpNode. test=develop
6 years ago
qingqing01 d8128930ef
Refine doc of uniform_random and fix dtype (#15873)
6 years ago
Xin Pan 44e7fcddc5
Merge pull request #15844 from panyx0718/infer
6 years ago
dzhwinter a71f2fbe4f fix default value. test=develop
6 years ago
Jacek Czaja dec9cf53c8 [MKL-DNN] MKL-DNN specific Tensor modification (#15429)
6 years ago
heqiaozhi 08c96d1b48 remove mkldnn & fix commit
6 years ago
minqiyang 84bf4d7b06 Move ClearBlock into OpBase and VarBase's destructor
6 years ago
minqiyang 2b3510bc50 Add imperative python tracer
6 years ago
minqiyang e9fdf9090d Polish code
6 years ago
Xin Pan 5dd281f738 polish
6 years ago
heqiaozhi 725b98f2c2 remove mkldnn & fix commit
6 years ago
peizhilin 6ccdb1b947 fix build issue on windows for sample prop op
6 years ago
heqiaozhi fab09ac0b8 Merge branch 'new_develop' of https://github.com/colourful-tree/Paddle into new_develop
6 years ago
heqiaozhi da4f5a2f18 remove mkl & fix commit
6 years ago
colourful-tree f2d6473ef8
Merge branch 'develop' into new_develop
6 years ago
heqiaozhi 04f876f5bc remove mkl & fix commit
6 years ago
dengkaipeng 373cfb0ccf use kernel size in global_pooling. test=develop
6 years ago
dengkaipeng 60305196b8 fix spell mistakes. test=develop
6 years ago
Dun c6bd434ffe
add memset CUPTI && test=develop (#15868)
6 years ago
乔龙飞 Qiao Longfei ec8e878200
Merge pull request #15840 from jacquesqiao/revert-15684-revert-15661-fix-cpu-broadcast
6 years ago
minqiyang a15a3fc314 Polish code
6 years ago
Tao Luo 8a7efc78f1
Merge pull request #15882 from sfraczek/unique_ptr_dereference
6 years ago
tensor-tang a0c37662b9 enable sgd jitkernel refer code and test
6 years ago
xuezhong 1dad36f6aa
Merge pull request #15609 from xuezhong/add_sample_logits_op
6 years ago
Qiao Longfei 2b7931d5c9 refine code test=develop
6 years ago
Kaipeng Deng 9e524a7b51
Merge pull request #15870 from heavengate/fix_adaptive_pool_doc
6 years ago
Xin Pan 8d83e38a6b remove mutex
6 years ago
shippingwang eb932f717a add cosine decay op, test=develop
6 years ago
Xin Pan 0362ef75f4 fix
6 years ago
dengkaipeng 14df92fe8f fix spell error. test=develop
6 years ago
dengkaipeng 144016fcfc fix adaptive_pool and yolov3_loss. test=develop
6 years ago
minqiyang 9dc64edfd9 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into imperative_engine
6 years ago
Xin Pan 12a0e2ed9d polish codes
6 years ago
Xin Pan 19d78f6797 polish
6 years ago
Sylwester Fraczek 74672d1aff Change *(smart_ptr.get()) -> *smart_ptr
6 years ago
tensor-tang ee2321debd
Revert 15770 develop a6910f900 gelu mkl opt (#15872)
6 years ago
xuezhong 81870723c6
Merge pull request #15605 from xuezhong/fix_bug_for_lstmp
6 years ago
dengkaipeng eb65b4e47d \frac -> \frac. test=develop
6 years ago
chengduo 3b08c9abf4
enhance profiler (#15842)
6 years ago
dengkaipeng 8167588f14 add blank after math::. test=develop
6 years ago
Xin Pan 32d5a16036 resolve conflicts
6 years ago
Qiao Longfei 3f9263f67e optimize style test=develop
6 years ago
dengkaipeng d9ec605873 use math:: instead of 29. test=develop
6 years ago
Qiao Longfei 4233d0a820 add more comment test=develop
6 years ago
dengkaipeng 19292ac6a1 fix adaptive pool doc.test=develop
6 years ago
Yiqun Liu 7d96c74ab2
Initialize the benchmark tester for operator. (#15772)
6 years ago
Yihua Xu 676995c86c Optimze Gelu with MKL Erf function (#15770)
6 years ago
Michał Gallus c4faf36e7a MKL-DNN: Add test for conv bias fuse pass (#15824)
6 years ago
Qiao Longfei 3bccc1e6e2 optimize broadcast logic test=develop
6 years ago
mozga-intel 5d132ecf83 Auto-cmake generator, auto-fill map (#15402)
6 years ago
Tao Luo 3831a4695d
Merge pull request #15862 from sfraczek/add-override-to-apply_impl
6 years ago
Tao Luo 075b5502b0
Merge pull request #15856 from kbinias/kbinias/activation-mkldnn-ut
6 years ago
Tao Luo 4774dad806
Merge pull request #15857 from sfraczek/fix-typo
6 years ago
Tao Luo e3dd6970fc disable dam temporarily (#15860)
6 years ago
Xin Pan 26e32e095a allow compiler to use graph
6 years ago
Sylwester Fraczek 1943119fc5 fix typo memeroy->memory
6 years ago
Sylwester Fraczek 0b926114c0 add override to ApplyImpl
6 years ago
minqiyang 8fe0c0c52c implement backward refs
6 years ago
Sylwester Fraczek 8bc604571f fix typo seriazlized->serialized
6 years ago
Sylwester Fraczek 543e53db05 fix typo releated->related
6 years ago
Krzysztof Binias 1578c60bdd Add new ut and remove unnecessary code
6 years ago
Dun Liang 35a90e06bf test=develop
6 years ago
Dun Liang c9080f516b test=develop
6 years ago
Dun Liang 1c7bb0e40c test=develop
6 years ago
Xin Pan 5eb87506bc add per kernel config and remove const_cast.
6 years ago
Qiao Longfei 62f1248ff5 fix use gpu test=develop
6 years ago
xuezhong a7e7d95203
Merge pull request #15630 from xuezhong/fix_bug_adagrad2
6 years ago
Xin Pan 6019054cdd
Merge pull request #15716 from Yancey1989/refine_pg
6 years ago