Commit Graph

765 Commits (3e1e1fee262fd16348fbd1e8f2e6231f791b3eeb)

Author SHA1 Message Date
Zeng Jinle 5eb381a3e2
refine reallocate of workspace size, test=develop (#19843)
6 years ago
Adam dfdd73cbc0 Add MKLDNNhandlerT templatized class (#19801)
6 years ago
Zeng Jinle 32b1151f5e
reduce default value of cudnn workspace size, test=develop (#19780)
6 years ago
Adam d4413a54bc Add common CreateKey for mkldnn handlers (#19767)
6 years ago
Yihua Xu 0d6ea52958 Fix the definition issue when used mkl_scsrmm and mkl_dcsrmm functions. (#19774)
6 years ago
Jacek Czaja 9e4c958552 Refactoring activation mkldnn op (#19748)
6 years ago
Huihuang Zheng 12542320c5
Replace TemporaryAllocator by CUDADeviceContextAllocator (#18989)
6 years ago
Adam 428b2b9e17 MKLDNN handler cleanup (#19713)
6 years ago
XiaoguangHu 27235cf222
Add document annotations for FLAGS that need to be open to external developers test=develop (#19692)
6 years ago
Tao Luo f05d2c519d paddle::framework::vectorize() templatization [PART3] (#19643)
6 years ago
Yiqun Liu 42b5bec6f9
Integrate NVRTC to support compiling CUDA kernel at runtime (#19422)
6 years ago
Tao Luo 3ae939e48a
unify PADDLE_ASSERT_MSG into PADDLE_ENFORCE(error_message) (#19631)
6 years ago
Tao Luo 75d1571995
refine PADDLE_ENFORCE codes for unify PADDLE_ASSERT_MSG (#19603)
6 years ago
Adam e94b26daf5 using MKLDNNMemoryFormat = mkldnn::memory::format changes (#19568)
6 years ago
Tao Luo 49523ea189
replace PADDLE_ASSERT with PADDLE_ASSERT_MSG (#19586)
6 years ago
zhouwei25 84c728013c fix the compilation issue on windows caused by mkl_CSRMM (#19533)
6 years ago
Jacek Czaja cef95ee30d [MKL-DNN] Refactoring Softmax (#19312)
6 years ago
Zeng Jinle 0a73f7202a
Add retry_allocator for gpu (#19409)
6 years ago
Jacek Czaja ecd9f330c9 [MKL-DNN] Fix to face model on AVX512 platforms (#19282)
6 years ago
liuwei1031 d6cb1a4122
add dynamic C runtime support on windows, test=develop (#19502)
6 years ago
Zeng Jinle c2c5b1b941
remove signal raise msg, test=develop (#19527)
6 years ago
Zeng Jinle caf59d0f3f
Add signal message to stderr (#19421)
6 years ago
Yi Liu efb05ba258
supports multiple NCCL communicators preserved in NCCLCommContext (#19407)
6 years ago
wopeizl b8aa37d529
save the callstack information to file when exception throws test=dev… (#19324)
6 years ago
Tao Luo 6527a7df67
replace part of PADDLE_ASSERT to PADDLE_ENFORCE (#19285)
6 years ago
Yihua Xu b920395842 Use sparse matrix to implement fused emb_seq_pool operator (#19064)
6 years ago
Zeng Jinle 91a0911ca3
Make PADDLE_ENFORCE_EQ support types that cannot be converted to std::string (#19243)
6 years ago
Zeng Jinle 708bd9798d
move_flags_to_unified_files_for_management, test=develop (#19224)
6 years ago
Zeng Jinle 002f325dcd
add PADDLE_ENFORCE_CUDA_SUCCESS, test=develop (#19211)
6 years ago
Adam b837689e97 Add generalized Conv+Activation MKLDNN fuse pass creation (#19072)
6 years ago
gongweibao 29d8781240
Polish fleet API to support cuda collective mode and nccl2 mode. (#18966)
6 years ago
wopeizl 80b7ef6fc8
add tensorrt support for windows (#19084)
6 years ago
Zhang Ting c2063217e7 optimize error message for "embedding" and "cross_entropy" OP (#18765)
6 years ago
liuwei1031 a43a763b54
fix warpctc.dll not found issue (#18761)
6 years ago
Zeng Jinle 08fa98f7cc
Fix gpu_info PADDLE_ENFORCE_GT when fraction_of_gpu_memory_to_use=1.0 (#18950)
6 years ago
Jacek Czaja 5cf2d38594 - Removed passing X from FWD to GRAD via device context (#18911)
6 years ago
Huihuang Zheng ea6ee76fa9
GPU allocation uses fraction of available memory (#18896)
6 years ago
Jacek Czaja cfcb96d2df [MKL-DNN] Fix int8 performance regression (#18758)
6 years ago
Huihuang Zheng 0d3f16f53e
Try to modify external gflags to solve CI compilation (#18872)
6 years ago
Huihuang Zheng cfce4994cf
Merge cuda 9/10 dockerfile with root dockerfile (#18693)
6 years ago
lidanqing 9ecd8ee789 change ComputeINT8 to template version to remove checking dst_datatype code (#18756)
6 years ago
Jacek Czaja 95c1816ec0 [MKL-DNN] Extended LRN with reusing via Acquire API (#18675)
6 years ago
chengduo fd3aad6cb3
Make fuse_optimizer_op_pass also work when the model contains sparse gradients. (#18664)
6 years ago
Jacek Czaja 0d8e6c9b8b MKL-DNN upgrade to 0.20 (#18370)
6 years ago
zhouwei25 772e09560e Optimize the content of error reporting information, print error code and official document web sites (#18671)
6 years ago
Zeng Jinle ae58afc546
Feature/auto_growth_allocator (#18561)
6 years ago
liuwei1031 759530966c
print out error code of cudaGetDeviceProperties if failed (#18643)
6 years ago
Jacek Czaja 71d883b8ef [MKL-DNN] Reimplemented pool2d mkl-dnn to use Acquire API (#18585)
6 years ago
Tao Luo 076f833110
add config.SetMkldnnCacheCapacity api for mkldnn cache clear strategy (#18580)
6 years ago
gongweibao c0a82748cf
Polish backwards optimizer dependency codes and use more default values. (#18255)
6 years ago