25032 Commits (d4413a54bc95e80d54403fd5c48261ca7313d125)

 

All Branches   

Search

Author	SHA1	Message	Date
Adam	d4413a54bc	Add common CreateKey for mkldnn handlers (#19767) ... test=develop	5 years ago
Yihua Xu	0d6ea52958	Fix the definition issue when used mkl_scsrmm and mkl_dcsrmm functions. (#19774) ... test=develop	5 years ago
chengduo	056fdedde3	Open fuse all reduce option (#19765) ... * Open fuse all reduce op test=develop * Add Fuse optimization op log * Add log in fuse_optimizer op pass and fuse all_reduce op pass * replace with boost::optional<bool> test=develop * Polish code test=develop * fix code coverage test=develop	5 years ago
Aurelius84	8c7e411908	Remove constraint that last dimension is forced to be 1 by adding one_hot_v2 (#19716) ... * add one_hot_v2_op to remove last_dims==1 test=develop * add api unittest code for CI_Coverage test=develop * improve CI_Coverage rate by adding test_with_depth test=develop	5 years ago
JesseyXujin	e352467c1c	modify activation op API, delete use_cudnn args, test=develop, (#19758)	5 years ago
Jacek Czaja	9e4c958552	Refactoring activation mkldnn op (#19748) ... test=develop - fix to BWD test=develop	5 years ago
Huihuang Zheng	12542320c5	Replace TemporaryAllocator by CUDADeviceContextAllocator (#18989) ... TemporaryAllocator is a singleton used for allocating memory for Cudnn. Since it is a singleton, we can delete it for better performance in memory. We replace TemporaryAllocator by CUDADeviceContextAllocator and CUDADeviceContextAllocation, which uses stream callback to delete the memory allocated for the stream to avoid singleton. Also added data_feed_proto to operator to fix CI in CPU compilation	5 years ago
Zeng Jinle	0daa5c9772	Make leaky relu inplacable (#19676) ... * make leaky relu inplacable, test=develop * force add unittests to pass coverage, test=develop	5 years ago
chengduo	c308c88d71	Fix test_parallel_executor_test_while_train (#19723) ... Fix test_parallel_executor_test_while_train	5 years ago
Zeng Jinle	078a678219	refine math_op_patch, test=develop (#19727)	5 years ago
chengduo	e506c99c20	Open fuse broadcast option (#18833) ... * fix vlog level and fuse option type test=develop	5 years ago
Jacek Czaja	47f670d58c	- Softmax mkl-dnn refactoring (#19615) ... test=develop - Cosmetic fixes test=develop	5 years ago
Yiqun Liu	a65c728e5d	Implement the GPU kernel of fc operator (#19687) ... * Refine the codes related to fc op. * Add GPU implementation for fc functor. * Apply fc_fuse_pass in GPU inference. test=develop * Change the cmake for fc op. * Change PADDLE_ENFORCE to PADDLE_ENFORCE_EQ. * Add an attribute to set the activation type in fc_op. * Enhance the unittest of fc_op. test=develop * Remove the declaration of FCOpGrad back to the header file. test=develop * Set default value for newly added arguments in test_fc_op. test=develop	6 years ago
Aurelius84	22301115d0	Remove constraint that last dimension is forced to be 1 in huber_loss op (#19562) ... * Remove constraint that last dimension is forced to be 1 in huber_loss test=develop * add y[rank-1] == 1 when x_rank=y_rank test=develop * modify into contain_unknown_dim test=develop	6 years ago
chengduo	5866a7a5fe	Enable fused_all_reduce_op_handle support GPU and CPU Gradients (#19418) ... * Enable fused_all_reduce_op_handle support GPU and CPU Gradients	6 years ago
Youwei Song	3e5fb6361b	fix api-doc error for dygraph and backward (#19721) ... * update dygraph api-doc and backward api-doc, test=develop * update dygraph api-doc and backward api-doc, update api.spec, test=develop * update dygraph api-doc and backward api-doc, update api.spec, test=develop * update API.spec, test=develop	6 years ago
Thunderbrook	bda92434db	remove trainer desc test in windows temporarily (#19753)	6 years ago
Youwei Song	810a33c8bd	fix dygraph partitial backward problem (#19625) ... * fix dygraph partitial backward problem, test=develop * add unittest, fix ClearGradient. test=develop * add filter and error in python side, test=develop * rebase develop, test=develop * bug fix for list equals in py3.5, test=develop * bug fix for list equals, test=develop	6 years ago
Tao Luo	ec9bc1bd9f	paddle::framework::vectorize() templatization (#19730) ... remove unused accuracy-diff warpctc-cudnn implementation test=develop	6 years ago
Bai Yifan	aa63d5ac6d	Make the distillation process not save teacher variables in PaddleSlim (#19633) ... * split teacher checkpoints with student checkpoints, test=develop * add unittest for graph.merge(), test=develop	6 years ago
Zeng Jinle	bb4f8dee83	add logs to left var memory size, test=develop (#19722)	6 years ago
Adam	428b2b9e17	MKLDNN handler cleanup (#19713) ... * MKLDNN handler cleanup * MKLDNN handler cleanup test=develop	6 years ago
chengduo	2c30e64b2f	increase timelimit test_pe_serexnext (#19702) ... test=develop	6 years ago
XiaoguangHu	27235cf222	Add document annotations for FLAGS that need to be open to external developers test=develop (#19692) ... Add document annotations for FLAGS that need to be open to external developers	6 years ago
Zeng Jinle	1c25c88aba	refine memory usage of some operators, test=develop (#19700)	6 years ago
baojun	87f13f7569	upgrade ngraph to support mkldnn v1.0 (#19689)	6 years ago
wangguanzhong	25dcd74d34	merge empty lod tensor, test=develop (#19228) ... * merge_empty_lod_tensor, test=develop * fix multiclass_nms, test=develop * refine API.spec, test=develop * add unittest case for fetch, test=develop * add lod tensor test, test=develop * return index for multiclass_nms, test=develop * add api for multiclass_nms2 * update API.spc, test=develop * refine api doc, test=develop * fix test_detection.py, test=develop * polish code, test=develop * add more unittest case, test=develop	6 years ago
yaoxuefeng	c6756ed225	fix instag op (#19591) ... * fix instag op * fix instag bug: Some tiny logical error, occurring when ins_tag (2nd input) is multiple. test=develop	6 years ago
gongweibao	6c2bc29cc0	Fix float16 optimizer. (#19682) ... Fix float16 optimizer	6 years ago
Zeng Jinle	713c05dd60	refine tensor.mutable_data, test=develop (#19680)	6 years ago
Chen Weihang	c78a4781bf	Fix train error when test_program.clone is executed after optimizer.minimize (#19397) ... * add prune when test_program.clone is executed after optimizer.minimize * add unittest, test=develop * add resnet and transformer test case, test=develop * add regularization for optimizer & program compare function, test=develop * add lstm unittest, test=develop * polish code based on review comment, test=develop * adapt to interface change in framework._prune, test=develop * update API.spec, test=develop	6 years ago
zhongpu	5f627488db	add kernel for unsqueeze_op and Add unsqueezed op test, test=develop (#19436) ... * add kernel for unsqueeze_op, test=develop * add kernel for unsqueeze_op, test=develop * add kernel for unsqueeze_op, test=develop	6 years ago
Zeng Jinle	a7691603a5	add gpu_allocator_try_time config, test=develop (#19675)	6 years ago
JesseyXujin	0b06db9413	delete transmission args in linear_chain_crf op (#19619) ... * delete args on linear_chain_crf_op doc * delete args on linear_chain_crf_op doc * delete args on linear_chain_crf_op doc * add code example * fix api doc * fix doc of crf * fix doc of crf * add test=develop * modify API.spec, test=develop	6 years ago
tianshuo78520a	fbc63a1f1f	Change Dianhai GithubID to fit new code merge rules(#19674)	6 years ago
Tao Luo	f05d2c519d	paddle::framework::vectorize() templatization [PART3] (#19643) ... * paddle::framework::vectorize() templatization test=develop * update pybind/imperative.cc test=develop * revert update on unsqueeze_op.cc and warpctc_cudnn_op.cu.cc test=develop	6 years ago
XiaoguangHu	103d09169d	Update README_cn.md to latest version v1.5.2 ... Update README_cn.md to latest version v1.5.2	6 years ago
XiaoguangHu	c50b0d6b79	Update README.md to latest version v1.5.2 ... Update README.md to latest version v1.5.2	6 years ago
hutuxian	1ca6ea0318	fix cmakelist deps (#19668) ... fix cmakelist deps: remove unnecessary deps and add proper op deps	6 years ago
Tao Luo	bcddbc78d4	remove -Wmaybe-uninitialized warning (#19653) ... * remove -Wmaybe-uninitialized warning test=develop * remove uninitialized op_handle_ in scale_loss_grad_op_handle.cc test=develop	6 years ago
Zeng Jinle	2db40d9f60	reduce thread num of retry_allocator_test,test=develop (#19638)	6 years ago
wangchaochaohu	4440d7ced0	test=develop cuda realization of label smooth op (#19175)	6 years ago
chengduo	31c5a5ee26	Remove linear_chain_crf_op.cu (#19645) ... test=develop	6 years ago
123malin	a25a716e87	Optimize fleet API: add input check for some interfaces (#18971) ... * fleet api add input check, test=develop	6 years ago
wangchaochaohu	ed8f44ea21	codegen for fused elementwise operation (#19520) ... * test=develop codegen for fused elementwise operation * fix test=develop	6 years ago
Aurelius84	25c0eb28cc	fix unittest random failed test=develop (#19654)	6 years ago
chengduo	5c4eb394e9	Make test_pe_seresnext serial (#19634) ... * make test_pe_seresnext serial test=develop * Increase test_pe_seresnext time limit on MAC test=develop	6 years ago
chengduo	ac34aac33f	Disable the test_parallel_executor_test_while_train temporarily (#19641) ... * Disable this unit test temporally test=develop * Split GPU and CPU test=develop	6 years ago
Chen Weihang	73daa3d6c0	Code Cleanup: delete three useless raw variables in Conv2D (#19644) ... * delete useless raw variables in Conv2D, test=develop * adjust the vars number in test_graph_wrapper to pass unittest, test=develop	6 years ago
Jie Fang	c6a598a276	init new amp, optimize inserting cast op for batchnorm (#18596) ... init new amp, optimize inserting cast op for batchnorm	6 years ago