Commit Graph

13567 Commits (ecc12fb43025022e3cc35e34607874420ca397e8)

Author SHA1 Message Date
qingqing01 943d972878
Fix analysis predictor when loading the persistable RAW type variable. (#15613)
6 years ago
liuwei1031 6e84eb131f expose peak gpu memory API to python test=develop (#15529)
6 years ago
dzhwinter 5cab99a686 fuck windows. rerun windows ci. test=develop
6 years ago
dzhwinter 9c9ad7d40b Merge remote-tracking branch 'origin/develop' into feature/ir_inplace_pass
6 years ago
jerrywgz 1743d1a58f
Merge pull request #15356 from jerrywgz/add_clip_op
6 years ago
dzhwinter 0a63234c85 follow comments. test=develop
6 years ago
mozga-intel 43a67a2662 Enable conv2d operator for a ngraph engine (#15269)
6 years ago
tensor-tang a6a1a92ef7
Merge pull request #15586 from tensor-tang/jit/cache
6 years ago
jerrywgz 4f18a9b87b test=develop
6 years ago
JiabinYang 2c133430f4 test=develop, fix no_avx exit
6 years ago
guoshengCS 5dfce93101 To make CUDA_LAUNCH_KERNEL_HELPER support large size.
6 years ago
Yan Chunwei e887d71958
fix ir debug config (#15571)
6 years ago
Yan Chunwei 897789b16e
fix save_inferece_model bug (#15365)
6 years ago
lidanqing-intel 4b3c6612a1 optimize density_prior_box_op.h for cpu
6 years ago
tensor-tang ba02ac4692
use mat attr and refine test (#15448)
6 years ago
JiabinYang be36e186f6 test=develop, refine wget issue
6 years ago
JiabinYang 9640736ad7 test=develop, refine wget issue
6 years ago
dzhwinter 9e87fbebb7 rerun windows ci. test=develop
6 years ago
dzhwinter 6f9904e99a rerun windows ci. test=develop
6 years ago
jerrywgz 4eb44380a6
Merge branch 'develop' into add_clip_op
6 years ago
Xin Pan 30cc8b7a92
Merge pull request #15554 from heavengate/yolo_loss_darknet
6 years ago
shanyi15 f96f166c8c test=develop, refine doc
6 years ago
Tao Luo 1a252f4be6
Merge pull request #15587 from luotao1/bert
6 years ago
shanyi15 b612709fec test=develop, refine doc for fast_install
6 years ago
JiabinYang ed7ae471d3 test=develop, fix mac python check error
6 years ago
JiabinYang a6af361278 Merge branch 'fast_install_1.3' of https://github.com/JiabinYang/Paddle into fast_install_1.3
6 years ago
Haihao Shen 1b8047b712 Add INT8 calibration support in Paddle package (#15569)
6 years ago
JiabinYang ff85734eed Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into fast_install_1.3
6 years ago
mozga-intel 312500dcb5 Enable pool2d operator for a ngraph engine (#15395)
6 years ago
Jiabin Yang b4c24f3f7c
Merge pull request #15575 from JiabinYang/feature/imperative
6 years ago
dzhwinter a52be7c081 refine build strategy. test=develop
6 years ago
Tao Luo ea92905be4
Merge pull request #15478 from kbinias/kbinias/seperate-folders-for-mkldnn
6 years ago
dzhwinter 32a2014939 refine build strategy. test=develop
6 years ago
Yibing Liu 170842cbb4
Some improvements to support bert mixed precision training (#15585)
6 years ago
Yiqun Liu 16d54f7f23
Return parent_idx in beam_search op (#15520)
6 years ago
jerrywgz 72ee3c6232
Merge pull request #15398 from jerrywgz/add_axis_for_boxcoder
6 years ago
jerrywgz e402c0ec7d test=develop
6 years ago
Kaipeng Deng d3eeb92bba
Merge pull request #15491 from tink2123/new_align_corners
6 years ago
jerrywgz 3046799ecd
Merge branch 'develop' into add_clip_op
6 years ago
dzhwinter 1a44b2fbe8 Merge remote-tracking branch 'origin/develop' into feature/ir_inplace_pass
6 years ago
Jiabin Yang 2d0ffdc485 test=develop, fix debug mode unitest, hsigmoid (#15574)
6 years ago
Zhaolong Xing 90ffe74954
Merge pull request #15546 from NHZlX/fix_trt_utest_random_failed
6 years ago
luotao1 8f0c2b07f2 use embedding=128 bert model for test
6 years ago
tensor-tang 2b0811c3fb refine vadd jitkernel choice
6 years ago
tensor-tang a18c0d4242 cache fc kernel
6 years ago
tensor-tang 6e1ee7fb57 cache softmax kernel func
6 years ago
Krzysztof Binias 69b7c595d6 Small fix
6 years ago
Krzysztof Binias b1bdcd4de8 Make separate folders for mkldnn codes
6 years ago
dzhwinter 06f2448848 Merge remote-tracking branch 'origin/develop' into feature/ir_inplace_pass
6 years ago
dzhwinter 8156fedf56 merge develop branch. test=develop
6 years ago
dengkaipeng 23d34d1f7e move yolov3_loss to detection. test=develop
6 years ago
JiabinYang 16f64b43d4 test=develop, Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/imperative
6 years ago
tensor-tang c7449227e8
Merge pull request #15563 from tensor-tang/jit/softmax
6 years ago
dengkaipeng 733bb82ec0 downsample -> downsample_ratio. test=develop
6 years ago
dengkaipeng ae0b0d5f93 fix doc. test=develop
6 years ago
dengkaipeng 56e21c558e add comments and docs. test=develop
6 years ago
dengkaipeng 577424e5ec use darknet loss and trick
6 years ago
dengkaipeng 042fecefab use L2Loss. test=develop
6 years ago
dengkaipeng af124dcdf6 fix API error
6 years ago
dengkaipeng c945ffa7f8 fix label_smooth and mixup score
6 years ago
tink2123 2b89f59055 add attr use_label_smooth test=develop
6 years ago
dengkaipeng 8218e30176 add gtscore. test=develop
6 years ago
dengkaipeng 3c08f620c2 add label smooth. test=develop
6 years ago
dengkaipeng cc01db6029 calc valid gt before loss calc. test=develop
6 years ago
dengkaipeng 32d533c2cd cache obj_mask and gt_match_mask. test=develop
6 years ago
dengkaipeng 6c5a5d0789 format code. test=develop
6 years ago
dengkaipeng e7e4f084e5 ignore pred overlap gt > 0.7. test=develop
6 years ago
dengkaipeng bd6deb1a8b fix API.spec change. test=develop
6 years ago
dengkaipeng db8ff57a61 remove useless code and update doc. test=develop
6 years ago
dengkaipeng 577a92d992 use typename DeviceContext. test=develop
6 years ago
dengkaipeng 0c4acc8305 imporve yolo loss implement. test=develop
6 years ago
dengkaipeng 2fbfef2ec9 fix no box expression. test=develop
6 years ago
dengkaipeng c0fa8d2eec use L1Loss for w, h. test=develop
6 years ago
dengkaipeng 3841983aa0 fix division error in mean process. test=develop
6 years ago
dengkaipeng 192d293854 use stable Sigmoid Cross Entropy implement. test=develop
6 years ago
Tao Luo 245b1f0579
Merge pull request #15570 from luotao1/bert
6 years ago
tink2123 909f864a9b remove unnecessary flags
6 years ago
JiabinYang bb881199f2 test=develop, polish code and fix wrong change in /paddle/fluid/inference/utils/CMakeLists.txt
6 years ago
tink2123 6961a94e94 avoid out_size less than 1
6 years ago
Jiabin Yang 075df09f86
Merge pull request #15470 from JiabinYang/feature/imperative
6 years ago
Qiyang Min b69996c2d3
Merge pull request #15558 from velconia/imperative_resnet
6 years ago
luotao1 5504425eb3 fix compiler error, use len20 dataset for bert
6 years ago
Yan Chunwei 655179089f
AnalysisConfig remove contrib namespace (#15540)
6 years ago
jerrywgz 7bc8481c62
Merge pull request #15418 from jerrywgz/refine_nms
6 years ago
tensor-tang d59f733551 refine softmax and use with cache
6 years ago
tensor-tang 7383eefd2d add softmax mix and mkl code
6 years ago
tensor-tang 50945685f2 add hmax, hsum jitcode
6 years ago
tensor-tang 8117725852 add jit kernel hsum, hmax and softmax refer code
6 years ago
Tao Luo 67e4450c34
Merge pull request #15485 from luotao1/fc500110-bert_test
6 years ago
Qiyang Min 6000a6e76e
Merge pull request #15312 from velconia/add_pyramid_dnn_support
6 years ago
Jiabin Yang fd286f3596
Merge pull request #15534 from JiabinYang/fix/multi_output_support_imperative
6 years ago
minqiyang 07822fef2c Clear all parameters' gradient
6 years ago
Zeng Jinle bf7dedcbc7
Merge pull request #15545 from sneaxiy/fix_debug_nccl_error
6 years ago
minqiyang 49a7fba848 Polish code
6 years ago
minqiyang 159c407328 Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into imperative_resnet
6 years ago
minqiyang 5c7768776c Fix batch_norm's stop_gradient bug
6 years ago
luotao1 e31aef9f6e Merge branch 'develop' into fc500110-bert_test
6 years ago
qingqing01 a6910f900e
Always create variables in analysis_predictor before OptimizeInferenceProgram. (#15533)
6 years ago
tink2123 e7eb08febe fix api.spec
6 years ago
Tao Luo 748c2d3ea2
Merge pull request #15530 from luotao1/remove_with_doc
6 years ago