mindspore2022

Commit Graph

Author	SHA1	Message	Date
mindspore-ci-bot	8945884137	!8990 close BatchMatmul and ReduceSum in graph kernel From: @looop5 Reviewed-by: @gaoxiong1,@ckey_dou Signed-off-by: @ckey_dou	5 years ago
mindspore-ci-bot	22d683a805	!8920 Adapt ops LinSpace for Ascend. From: @liu_xiao_93 Reviewed-by: @liangchenghui,@linqingke,@liangchenghui Signed-off-by: @liangchenghui,@liangchenghui	5 years ago
mindspore-ci-bot	ddff3c4277	!8969 [bug_fix]GPU distributed training occur core dump when memory is not enough From: @zyli2020 Reviewed-by: @limingqi107,@cristoval Signed-off-by: @cristoval	5 years ago
mindspore-ci-bot	42cbdfcafc	!8903 add trace info when mindspore error From: @jjfeing Reviewed-by: @chujinjin,@zh_qh,@kisnwang Signed-off-by: @zh_qh	5 years ago
looop5	1b36f454b8	close BatchMatmul and ReduceSum in graph kernel	5 years ago
liuxiao93	584e241e29	Adapt ops LinSpace for Ascend.	5 years ago
mindspore-ci-bot	08dc1481c7	!8918 checkcircle should consider the edges coming from controldepent nodes From: @lingyunli63 Reviewed-by: Signed-off-by:	5 years ago
mindspore-ci-bot	661b6073a4	!8921 fix return scalar From: @jjfeing Reviewed-by: @kisnwang,@chujinjin Signed-off-by: @chujinjin	5 years ago
jjfeing	27257b9901	add trace when mindspore error	5 years ago
lizhenyu	6f6a0dfd7a	[bug_fix]GPU distributed training occur core dump when memory is not enough	5 years ago
lingyunli63	e6a5fc0739	consider controldepend edges in checkcircle	5 years ago
jjfeing	a607890256	fix_return_scalar	5 years ago
mindspore-ci-bot	3f75f13556	!8648 PyNative Performance Optimization From: @jojobugfree Reviewed-by: Signed-off-by:	5 years ago
caifubi	c7d6997819	pynative host device parallel	5 years ago
mindspore-ci-bot	420a4dc162	!8857 fix bug of change axis of reduce kernel when format is 6hd or fracz From: @lianliguang Reviewed-by: @kisnwang,@jjfeing Signed-off-by: @jjfeing	5 years ago
lizhenyu	094f0b2a07	bugfix:fused batch norm op's input channel nums should be a multiple of 4	5 years ago
LianLiguang	29d585385e	fix bug of change axis reduce	5 years ago
laiyongqiang	978b7e2e18	fix codex and review bot warning	5 years ago
fangzehua	69ce58425d	fix reshape dynamic and emb	5 years ago
mindspore-ci-bot	fb0e866ad1	!8269 forward unique dynamic shape From: @yao_yf Reviewed-by: Signed-off-by:	5 years ago
yao_yf	31819bb4a7	support forward unique	5 years ago
mindspore-ci-bot	286f5b05f7	!8493 【GraphKernel】Fuse composite ops separated by GetItem nodes From: @dayschan Reviewed-by: @ckey_dou Signed-off-by: @ckey_dou	5 years ago
mindspore-ci-bot	9969c83f75	!8689 [GraphKernel] Split shape ops for more fusion opportunity. From: @tronzhang Reviewed-by: @gaoxiong1,@ckey_dou Signed-off-by: @ckey_dou	5 years ago
liuxiao93	2aaf5e2e1b	Adapt DynamicGRUV2Grad for Ascend new backend.	5 years ago
dayschan	8e6d92eac9	Fuse composite ops separated by GetItem nodes	5 years ago
mindspore-ci-bot	381455638a	!8600 make gaps not lifelong: constraints from neighbor From: @laiyongqiang Reviewed-by: Signed-off-by:	5 years ago
mindspore-ci-bot	05f858e3d6	!8597 remove semi-lifelong for communication op's input's memory From: @laiyongqiang Reviewed-by: @kisnwang Signed-off-by:	5 years ago
mindspore-ci-bot	dcff06a5d6	!8576 【GraphKernel】Disable all simplify pattern except SimplifyReduce in arithmetic_simplify. From: @dayschan Reviewed-by: Signed-off-by:	5 years ago
tronzhang	9d7494f4df	split shape ops for more fusion pportunity.	5 years ago
mindspore-ci-bot	66c6fe3d6a	!8603 handle getnext output tensor as normal lifelong tensor From: @laiyongqiang Reviewed-by: @kisnwang Signed-off-by:	5 years ago
mindspore-ci-bot	07633f2a2a	!8601 independent node use the somas's memory From: @laiyongqiang Reviewed-by: @kisnwang,@jjfeing Signed-off-by: @jjfeing	5 years ago
dayschan	a8bb28437c	Temporarily disable all simplify pattern except SimplifyReduce, for some bizarre errors occurs in arithmetic simplify.	5 years ago
liubuyu	5bf70b24bb	adjust dynamic_rnn_grad_fission_v2 position	5 years ago
mindspore-ci-bot	b411c05de0	!8495 Adapt DynamicGRUV2 forward for Ascend new backend. From: @liu_xiao_93 Reviewed-by: Signed-off-by:	5 years ago
mindspore-ci-bot	df1d5333db	!8519 Migrate current graph kernel passes to Ascend back-end From: @looop5 Reviewed-by: Signed-off-by:	5 years ago
mindspore-ci-bot	24d04b1cb1	!8578 dynamic shape check From: @wilfchen Reviewed-by: @limingqi107,@cristoval Signed-off-by: @cristoval	5 years ago
lingyunli63	a51465c78b	add graphkerneloptimize pass align fuse_ops_fusion align composite_ops_fusion unify ops table Init new_code's kernel_info with orig_node's kernel_info in function NewCNodeWithInfo enable run bert add pass tensor_promotion add macro for bias_add and bias_add_grad in expander pass exclude unused attrs in primitive compare for GraphKernelCSE exclude fusion_type in kernelinfo cmp for cse in graphkernel check processor remove graph kernel pass before select kernel recover run_standalone_pretrain_ascend.sh remove is_before_kernel_select move add_atomic_clean from pass directory to graph_kernel directory update fuse op list in Ascend back-end	5 years ago
Ioannis Lamprou	7e72605fc8	make gaps not lifelong: constraints from neighbor	5 years ago
laiyongqiang	b8821bb2f3	handle getnext output tensor as normal lifelong tensor	5 years ago
laiyongqiang	222159599a	independent node use the somas's memory	5 years ago
laiyongqiang	42c3830938	remove lifelong for communication output	5 years ago
mindspore-ci-bot	3c6b91f6fa	!8548 handle getNextOutput and contiguous conflict From: @laiyongqiang Reviewed-by: @kisnwang,@jjfeing Signed-off-by: @jjfeing	5 years ago
wilfChen	2291b7f2e6	dynamic shape check	5 years ago
liuxiao93	d471ac491e	Adapt DynamicGRUV2 forward for Ascend new backend.	5 years ago
laiyongqiang	0fdc6b547d	handle getNextOutput and contiguous conflict	5 years ago
mindspore-ci-bot	dbe5229c56	!8492 expand maxmium_grad minimum_grad and dropout_grad From: @zengzitao Reviewed-by: @gaoxiong1,@ckey_dou Signed-off-by: @ckey_dou	5 years ago
mindspore-ci-bot	e65c68a723	!8475 DynamicRNNGrad run error: cast Int32Imm From: @gaojing22 Reviewed-by: @yingjy Signed-off-by: @yingjy	5 years ago
zengzitao	28f1db74dd	expand maximum_grad minimum_grad dropout_grad op	5 years ago
mindspore-ci-bot	7b70c17fc0	!8449 【GraphKernel】Add Transpose into fusible list; Update akg submodule. From: @dayschan Reviewed-by: @gaoxiong1,@ckey_dou Signed-off-by: @ckey_dou	5 years ago
mindspore-ci-bot	b078954667	!8389 remove multiple circles From: @lingyunli63 Reviewed-by: @gaoxiong1,@ckey_dou Signed-off-by: @ckey_dou	5 years ago

1 2 3 4 5 ...

335 Commits (a17f76dd1d83728cdc8ffcb52de694dfd3fcf12e)