Paddle

Commit Graph

Author	SHA1	Message	Date
chengduo	8259f1418f	Enhance backward process (#18700 ) * prun backward ops test=develop	7 years ago
chengduo	e0d8c6ac68	Add find_no_grad_vars in backward.py (#17942 ) * add not_been_used_vars to no_grad_set test=develop	7 years ago
xsrobin	47e2ef38e9	add "import paddle.fluid as fluid" to examples lack of it	7 years ago
qingqing01	80d2e66f9e	Update backward appending stragety to support double backward and fix some bug. (#18104 ) * Update backward.py: - If there is no input grad var in all outputs of previous ops, do not append this op into graph. - Only apply this stragety when double backward. * Update some double backward op. * Update sum_op to judge whether a tensor is empty by numel or IsInitialized().	7 years ago
Zeng Jinle	712bfb17cb	fix recurrent_op,test=develop (#17433 )	7 years ago
lujun	e388a1fb66	Repair api example (#17221 ) Fix the following API examples: paddle.fluid.scope_guard paddle.fluid.backward.append_backward paddle.fluid.cpu_places paddle.fluid.cuda_pinned_places paddle.fluid.cuda_places paddle.fluid.in_dygraph_mode paddle.fluid.CUDAPlace paddle.fluid.CPUPlace paddle.fluid.CUDAPinnedPlace	7 years ago
qingqing01	c1c2633a63	Support backward of backward for Relu and add a new gradient checker by comparing theoretical and numerical Jacobian. (#16862 ) * Support backward of backward and a new gradient checker * Rename decorators.py to decorator_helper.py, since Python on Windows CI has decorators package. 1. Add ReluDoubleGradMaker when register relu_grad. 2. Add a new gradient checker by comparing theoretical and numerical Jacobian. Check double gradients by double_grad_check.	7 years ago
Zeng Jinle	1c526e1d1a	Fix some grad op desc makers (#16633 ) * fix some grad op desc maker test=develop * fix grad op desc makers test=develop	7 years ago
Xin Pan	6324032602	MLP forward backward test=develop	8 years ago
Xin Pan	47ea2534fb	clean parallel do test=develop	8 years ago
Wu Yi	16e73e0d28	hide operator API (#12543 ) * hide operator API * update * update api.spec * fix merge * fix test	8 years ago
Wu Yi	efafc72f62	Hide program APIs (#12315 ) * hide program APIs * fix merge error * update	8 years ago
minqiyang	99d3f08920	Add print_function for all python files	8 years ago
minqiyang	77f12e000f	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11	8 years ago
gongweibao	842fb021b3	Fix clone() bug. (#12583 )	8 years ago
minqiyang	e0d5f8a820	Move compat module to python/paddle	8 years ago
minqiyang	5338417b47	Polish code style	8 years ago
minqiyang	ae39709e59	Polish code	8 years ago
minqiyang	e4057d071b	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_pybind11	8 years ago
minqiyang	5d4238cdcc	Fix six.iteritems problem	8 years ago
minqiyang	6dc07e7f95	Replace items() with six.moves.iteritems() to improve memory usage	8 years ago
minqiyang	bc12c2c616	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into port_python3_syntax	8 years ago
minqiyang	c3fdf3aee4	Fix divide problem in CI Fix pb_protobuf2 FromString problem	8 years ago
minqiyang	6abe819f07	Fix pybind11 problem Fix str and bytes problem Fix sorted problem Fix math problem Fix CI problem	8 years ago
Yu Yang	55ff03b7ab	Do not set loss@Grad as persistable Revert part of `a3ca4c99eb`	8 years ago
minqiyang	559d36328c	Apply 2to3 to current paddle main python code	8 years ago
Wu Yi	db67d60e31	Remove block api (#12107 ) * remove block api * remove clone_variable * hide block inner apis * update * fix tests	8 years ago
Yu Yang	1c9c8e8de7	Merge pull request #12065 from reyoung/feature/hide_apis Add deprecated annotation and hide a lot of APIs	8 years ago
yuyang18	ff07af8d93	Hide calc_gradient	8 years ago
fengjiayi	07ab9ce4e7	update the comments	8 years ago
fengjiayi	3c9b59b8e8	update	8 years ago
fengjiayi	f5894d22c5	Fix a backward bug	8 years ago
tensor-tang	d5fb8fa778	Revert "Merge pull request #11628 from PaddlePaddle/revert-11102-mozga-intel/Sum_mkldnn_layout" This reverts commit `4d8e8ee226`, reversing changes made to `d6a9f005c8`.	8 years ago
tensor-tang	90780e22ce	Revert "MKLDNN layout: Support for sum operator"	8 years ago
Yu Yang	c2ad3815b0	Merge pull request #11560 from JiayiFeng/doc_non_layer_api Doc of non layer api	8 years ago
fengjiayi	457d81bbc0	fix errors	8 years ago
mozga-intel	96b4904d2f	MKLDNN layout: Support for sum operator	8 years ago
fengjiayi	8ea54e2f95	Add docs	8 years ago
yuyang18	23e19e2e42	Fix bug	8 years ago
yuyang18	fb370f4411	Refine code	8 years ago
yuyang18	3923d40992	Merge branch 'develop' of https://github.com/PaddlePaddle/Paddle into feature/support_op_role	8 years ago
yuyang18	44c52a8c1a	Polish op_proto_maker	8 years ago
yuyang18	017bba1664	Add op role	8 years ago
chengduoZH	a3ca4c99eb	fix loss.gradVar	8 years ago
chengduoZH	ef28e7deba	refine parallel_do_grad	8 years ago
QI JUN	f7e9fe57d3	[Memory]More memory optimization policy (#8690 ) * add memopt level * add opt level for image classification demo * clean code * add delete op * clean code * test machine translation demo * clean code * clean code * skip fill constant with force cpu * clean code * clean code * refine code * clean code * fix bug	8 years ago
fengjiayi	55e11e21cb	fix_a_bug	8 years ago
qiaolongfei	f8d0d84f7e	fix multi card	8 years ago
Luo Tao	bde090a975	replace paddle.v2.fluid by paddle.fluid in tests	8 years ago
Luo Tao	b11956a0b5	move Fluid API code out of V2 API code	8 years ago

50 Commits (c2c876f718b5fe77a55e43cf397dfd85e898dddb)