1214 Commits (b854d959a543ee83e89a77d0627fb375bf0f9ba1)

Author	SHA1	Message	Date
Xin Pan	7fb5b66ac2	Merge pull request #13916 from panyx0718/fix2 ... Make Var::GetMutable robust	6 years ago
Yan Chunwei	6809238d97	fix analysis predictor profile (#13896)	6 years ago
Xin Pan	342e436158	Make Var::GetMutable robust ... test=develop	6 years ago
Qiyang Min	e3a64fca44	Merge pull request #13835 from velconia/fix_reshape_op ... Fix Reshape op when input is the same with output	6 years ago
Zeng Jinle	af91d41ab8	Merge pull request #13852 from sneaxiy/feature/eager_delete_tensor ... Fix bug of eager deletion to support if_else_op	6 years ago
Xin Pan	288a112ffd	Revert "Revert "Revert "Make variable::GetMutable robust"""	6 years ago
Tao Luo	c26f2b21eb	Merge pull request #13813 from sfraczek/sfraczek/conv-bn-fuse-pass-full-eigen ... rewrite conv_bn fuse pass fully to eigen	6 years ago
Sylwester Fraczek	50c5e9b0c6	reshape_2d used from ddim.h ... test=develop	6 years ago
minqiyang	aeec82acd5	Add unittest for reshape op ... test=develop	6 years ago
Xin Pan	ddb76d0d09	Make GetMutable more robust ... test=develop	6 years ago
minqiyang	24c9fbdba3	Polish code ... test=develop	6 years ago
sneaxiy	d3ed070e10	test=develop	6 years ago
minqiyang	d9b202e717	Move tensor copy src_ptr and dst_ptr check to TensorCopy function ... test=develop	6 years ago
sneaxiy	fb6201e93e	test=develop	6 years ago
Xin Pan	d852be7c48	Revert "Make variable::GetMutable robust"	6 years ago
Xin Pan	ae8b1c32bf	Merge pull request #13821 from panyx0718/fix ... Make variable::GetMutable robust	6 years ago
Sylwester Fraczek	55d6950a1a	rewrite conv_bn fuse pass to eigen ... test=develop	6 years ago
Xin Pan	228506618b	Avoid GetMutable implicitly reset Var Type. ... This can cause a lot of problem: 1. Wrong operator implementation, Op can get a wrong type without failure. 2. Anytype can be Get without defined in VarType. Also fix wrong STEP_SCOPE usage. test=develop	6 years ago
Tao Luo	9b11a17502	Revert "[MKLDNN] Pass: Fuse Conv + Bias"	6 years ago
Tao Luo	ce248a15d9	Merge pull request #13368 from Sand3r-/mgallus/conv-bias-pass ... [MKLDNN] Pass: Fuse Conv + Bias	6 years ago
Tao Luo	16b1beb244	Merge pull request #13486 from sfraczek/sfraczek/conv-bn-fuse-pass ... Sfraczek/conv bn fuse pass	6 years ago
Michal Gallus	40b17be4b0	Pass: Fuse Conv + Bias ... test=develop	6 years ago
Sylwester Fraczek	3fcca40909	eigen sqrt fix and change 1e-5 to epsilon ... test=develop	6 years ago
chengduo	e1761709f8	Set the right shape of selected_rows (#13723) ... * set the right shape of selected_rows test=develop * enhance check * fix activation_op * remove cast * use ShareDimInfo replace SetDim and ShareLod * use ShareDimAndLod test=develop * follow comment test=develop * check whether the input has lod test=develop * Split ShareDimAndLod test=develop * checkout clip.py test=develop	6 years ago
dzhwinter	a46e30aa6d	enhance isinf/isnan in tensor util, avoid copy back to cpu (#12688) ... * "avoid copy back to cpu" * "add infinity support" * "fix ci" * "add cpu macro" * rerun ci; test=develop * "fix api" test=develop * test=develop * test=develop * test=develop * test=develop * test=develop	6 years ago
Xin Pan	943e4deb23	Merge pull request #13750 from panyx0718/fix ... clean unused code and small optimize	6 years ago
Sylwester Fraczek	78f98294c2	conv bn fuse pass ... review fix review from hshen14 fix test=develop fix error in broadcast and code cleanup rename bias -> eltwise and added macro to shorten code formatting	6 years ago
Xin Pan	d2079b1ddb	clean unused code and small optimize ... test=develop	6 years ago
Xin Pan	ab798a2832	clarify the fraction_of_gpu_memory flag ... test=develop	6 years ago
Tao Luo	d770b9bda3	Merge pull request #13663 from luotao1/resnet50_ut ... add resnet50 inference unit-test	6 years ago
dzhwinter	32c260cd1f	"fix operators cmake" (#13581) ... * "fix operators cmake" * "rerun ci" test=develop	6 years ago
Tao Luo	6ef6a9180a	Merge pull request #13727 from Sand3r-/mgallus/enable-mkldnn-naive-exe ... Enable MKL-DNN in Naive Executor and Analysis Predictor	6 years ago
Michal Gallus	09d9d77a8f	Enable MKLDNN in Naive Executor ... test=develop	6 years ago
sneaxiy	9606b37ce4	test=develop	6 years ago
Yu Yang	186b2b138d	Revert "Merge pull request #13201 from reyoung/revert_callstack" (#13697) ... This reverts commit 21bb9e91fc, reversing changes made to 3fa68dc101. test=develop	6 years ago
Tao Luo	a89afd4c22	Merge pull request #13685 from luotao1/naive_cmake ... update libpaddle_fluid.a/so	6 years ago
chengduo	d6747a9ac2	make check_graph choosable (#13674) ... test=develop	6 years ago
Xin Pan	425a882165	Merge pull request #13643 from panyx0718/ir2 ... clean up channel	6 years ago
luotao1	a989a4e7c2	refine paddle_inference_helper.h	6 years ago
Xin Pan	642905958a	fix compile error ... test=develop	6 years ago
tensor-tang	ede4b230be	Merge pull request #13553 from jczaja/prv-fused_embedding_fc_lstm_op ... Adding fused_embedding_fc_lstm op	6 years ago
Xin Pan	6746b1fdf3	add missing header ... test=develop	6 years ago
Xin Pan	ddd60581b7	clean up channel ... test=develop	6 years ago
Tao Luo	83ca657f96	Merge branch 'develop' into resnet50_ut	6 years ago
Jacek Czaja	e202f33aa9	- Yet another clarification to comment ... test=develop	6 years ago
Tao Luo	21ee30595b	clean some CMakeLists ... test=develop	6 years ago
Jacek Czaja	1df69f7c9d	- Fix to comment ... test=develop	6 years ago
Xin Pan	d24f1f0aa4	Current scope needs to be thread-safe for training ... scope's API modifies its internal state. And scope's API can be called from multiple threads during traing. Hence, we need locks to protect the scope's internal states. We can optimize it in the future. But the current solution is buggy. test=develop	6 years ago
Yan Chunwei	c8744d118d	fea/infer executor and concurrency performance issue bug fix (#13451) ... - add naive executor - fix concurrency performance issue	6 years ago
chengduo	5175b3cb2b	Add GraphChecker (#13580) ... * add GraphNum test=develop * add graph number check in parallelExecutor test=develop * fix transformer_model bug test=develop * fix graph num	6 years ago