476066dc82
!14263 tensorprint adapt to print scalar
...
From: @yepei6
Reviewed-by: @kingxian,@kisnwang
Signed-off-by: @kingxian
4 years ago
381cbaa764
tf models support running fp16 mode and add models to the entrance guard
4 years ago
4182c1f02a
delete depend format for its order use
4 years ago
eb861aa93e
!14112 [MSLITE] Fix the bug of matmul int8 for arm32
...
From: @zhanyuan1
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
4 years ago
b13a8e407e
fix-some-api-bug
4 years ago
ecb27493b3
!14031 fix model buf may double free
...
From: @xutianchun
Reviewed-by: @HilbertDavid,@zhanghaibo5
Signed-off-by: @HilbertDavid
4 years ago
ac674a41b0
split file list
4 years ago
c30daa0ee8
fix fc &&matmul
4 years ago
7149e8c2c9
!14045 [Graph Kernel] add compare test case
...
From: @zengzitao
Reviewed-by: @gaoxiong1
Signed-off-by:
4 years ago
15ed2daa31
!14255 [GraphKernel] open Assign fusion and fix OptimizeAssign
...
From: @lingyunli63
Reviewed-by: @gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
e644a66973
!14237 [GraphKernel] infer processor from ms_context inside function CreateCNode
...
From: @looop5
Reviewed-by: @gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
d702e52c74
!13534 support launch kernel for graph mode
...
From: @kisnwang
Reviewed-by:
Signed-off-by:
4 years ago
6646c4f35d
!14251 rm space of BatchMatMul,update examples of BNTrainingUpdate and NLLLoss
...
From: @mind-lh
Reviewed-by: @ljl0711,@wuxuejian
Signed-off-by: @wuxuejian
4 years ago
bda7fc5153
add reshepe type for graph cache
4 years ago
45932bf73e
[MSLITE] support conv1d.
4 years ago
d52da91124
!14094 Add validation and modify the example code in the comment. The code and output result are wrong.
...
From: @lijiaqi0612
Reviewed-by:
Signed-off-by:
4 years ago
95f9f2c092
!14244 optimize exception mode when use undefined name in if for and while
...
From: @zhangbuxue
Reviewed-by: @ginfung,@zh_qh
Signed-off-by: @zh_qh
4 years ago
b6ef88c168
add stop_dataset_profiler option
4 years ago
2ec60543cd
tensorprint adapt to print scalar
4 years ago
7f4994af7c
!14186 Support while bprop
...
From: @liangzelang
Reviewed-by: @kisnwang,@jjfeing
Signed-off-by: @jjfeing
4 years ago
3fe2ea6f6b
!14233 fix mindrt cpplint warning
...
From: @wangcong666
Reviewed-by: @kisnwang,@limingqi107
Signed-off-by: @limingqi107
4 years ago
4574fe6162
!14204 Add origin_shape for async data dump
...
From: @jojobugfree
Reviewed-by: @chujinjin,@kisnwang
Signed-off-by: @kisnwang
4 years ago
932027253f
fuse Assign and fix optimizeAssign
4 years ago
d830a2b55a
optimize exception mode when use undefined name in if for and while
4 years ago
a06234ed13
!13834 [GraphKernel] Remove two graph kernel frontend pass
...
From: @tronzhang
Reviewed-by: @gaoxiong1,@gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
203106c133
!14198 [GraphKernel] Refactor GraphKernelExpander (4th submission)
...
From: @dayschan
Reviewed-by: @gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
af1e25252d
optimize fp32 prelu
4 years ago
2516e6cc23
rm space of BatchMatMul,update examples of BNTrainingUpdate and NLLLoss
4 years ago
3c9d8cb073
The input and output of batchnorm reverse operator increase pass in ascend platform under the mode of pynitve
4 years ago
bca8389e31
!14209 [MS][LITE]optimize anf export
...
From: @mengyuanli
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
4 years ago
c784e45daa
!14220 gpu add ci model
...
From: @yeyunpeng2020
Reviewed-by: @ddwsky,@hangangqiang
Signed-off-by: @ddwsky
4 years ago
f1af028bc5
remove wrong comment of laplace op
4 years ago
82aeb61e95
fix name of tensor from StringToTensor
4 years ago
0a0dc05d51
!14144 update sponge
...
From: @zhangxinfeng3
Reviewed-by: @wang_zi_dong,@ljl0711
Signed-off-by: @ljl0711
4 years ago
550876a2cb
!14187 add platform for connect_network_with_dataset
...
From: @wangnan39
Reviewed-by: @zh_qh,@zhunaipan
Signed-off-by: @zh_qh
4 years ago
dd43972839
infer processor from ms_context inside function CreateCNode
4 years ago
f9fa97cd0c
bugs
4 years ago
6c5361fb5c
!14131 fix cmsis gesture of pooling opcoder
...
From: @zoloft
Reviewed-by: @wangchengyuan
Signed-off-by: @wangchengyuan
4 years ago
72c6dad4ba
add compare_test case in gpu ci and update akg submodule
4 years ago
72bad339e7
!14065 New add grad accumulation for network.
...
From: @linqingke
Reviewed-by: @guoqi1024,@xu-yfei
Signed-off-by: @guoqi1024
4 years ago
10a179f78e
optimize anf export
4 years ago
3dd7e28fdd
!14202 remove dup act
...
From: @zhaozhenlong
Reviewed-by: @hangangqiang,@zhang_xue_tong
Signed-off-by: @hangangqiang
4 years ago
4610ea8308
!14163 fix executor last task not clear
...
From: @kisnwang
Reviewed-by: @jjfeing,@zhoufeng54
Signed-off-by: @zhoufeng54
4 years ago
f3d5a45245
!14146 [MS][LITE][STABLE]free global infershape buffer when program exit
...
From: @jpc_chenjianping
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
4 years ago
477009aede
!13077 Gpu inference config
...
From: @wilfchen
Reviewed-by: @limingqi107,@cristoval
Signed-off-by: @cristoval
4 years ago
f3297e021c
!14216 add the impl of graph scheduler
...
From: @limingqi107
Reviewed-by: @cristoval,@wilfchen
Signed-off-by: @wilfchen
4 years ago
3c77f75c6f
!14218 [MSLITE][DEVELOP] update lite runtime date in Mate40
...
From: @yangruoqi713
Reviewed-by: @zhang_xue_tong,@zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong,@zhang_xue_tong
4 years ago
0c6e2b950d
add gpu ci model
4 years ago
b31f102e47
!14203 npu do weight quant
...
From: @zhaozhenlong
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
4 years ago
64edb371bf
!14118 fix gpu shape bug
...
From: @yeyunpeng2020
Reviewed-by:
Signed-off-by: @ddwsky
4 years ago
25fce5de2d
[MSLITE][DEVELOP] update lite runtime date for version r1.2 in Mate40（Hisilicon Kirin9000）
4 years ago
66265198e9
add shard strategy validation
4 years ago
7b48d059c5
new add grad accumulation for network.
4 years ago
87bf1ec80f
delete mark_interface_fusion and tensor reuse frontend pass for graph kernel
4 years ago
24634feb63
!14018 Add nn.Conv3d and nn.Conv3dTranspose.
...
From: @liu_xiao_93
Reviewed-by:
Signed-off-by:
4 years ago
56c4145cc5
Refactor GraphKernelExpander (4th submission)
...
Decoupled the process of expanding node from GraphKernelExpander pass,
so that it can be rolled back when error occurs.
By the way, supported controlling cluster ops by flags.
4 years ago
db2668d72a
gpu inference context
4 years ago
1892a629c8
!14138 Update case for interface forwardvalueandgrad
...
From: @joylvliang
Reviewed-by: @ginfung,@zhunaipan
Signed-off-by: @ginfung
4 years ago
33e010db69
add the impl of graph scheduler
4 years ago
bfc51f96ce
support launch kernel for graph mode
4 years ago
aa5da5ecc6
updata stm32f746 readme && fix pooling opcoders
4 years ago
d9d3aaed19
!14141 numpy-native fix docstrings for select
...
From: @jachua
Reviewed-by: @guoqi1024,@zhunaipan
Signed-off-by: @zhunaipan
4 years ago
4b4b78eac8
!13966 Add test cases for explainer
...
From: @lixiaohui33
Reviewed-by:
Signed-off-by:
4 years ago
dc96d61eca
add origin_shape for dump
4 years ago
e9ada9fd1d
!14192 add the force transform to avoid the utf8 error
...
From: @yepei6
Reviewed-by: @kingxian,@kisnwang
Signed-off-by: @kingxian
4 years ago
ad140a8bf4
!14084 [GraphKernel] support matmul on D
...
From: @lingyunli63
Reviewed-by:
Signed-off-by:
4 years ago
66028a76e5
!14184 update graphegine commit id to master
...
From: @shenwei41
Reviewed-by: @lilongfei15,@xsmq
Signed-off-by: @xsmq
4 years ago
e9d57490c5
!14155 [MS][RDR] fix fps degradation of gpu trainning in master branch
...
From: @louie5
Reviewed-by: @ouwenchang,@lixiaohui33
Signed-off-by: @lixiaohui33
4 years ago
766d402ec2
npu weight quant support
4 years ago
4e6b73937a
update spongeQm
4 years ago
12ea873ee0
fix gpu shape bug
4 years ago
de856a1bac
md fix gpu core dump problem
4 years ago
ce3597b727
add the force transform to avoid utf8 error
4 years ago
7999e64e34
!14169 fix some errors in docs of probability
...
From: @bingyaweng
Reviewed-by: @wang_zi_dong,@sunnybeike
Signed-off-by: @sunnybeike
4 years ago
48cea2f100
!13595 [VM][Backend]Fix bug of create tuple parameter
...
From: @chenfei52
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
4 years ago
6ee5bee606
remove dup act npu pass
4 years ago
cba92f7ea1
add platform for connect_network_with_dataset
4 years ago
ba65fb9f3c
Support non-tail recursive graphs
4 years ago
d9f6a6277d
!14149 Unfold repeated label index in labelswitch
...
From: @liangzelang
Reviewed-by: @xsmq,@lilongfei15
Signed-off-by: @lilongfei15
4 years ago
0625209cd1
!14002 CUDNN STATUS NOT SUPPORTED
...
From: @ding_fei_fei
Reviewed-by: @zhunaipan,@kingxian
Signed-off-by: @kingxian
4 years ago
e3b2b1e50a
update graphegine commit id to master
4 years ago
4b966ed40d
support matmul on D
4 years ago
a1621dbb77
!14159 Update API document example of ScatterNDUpdate
...
From: @yuyiyang_3418
Reviewed-by: @wuxuejian,@ljl0711
Signed-off-by: @wuxuejian
4 years ago
31802d9759
!14119 tensor-rt library dynamic load
...
From: @wilfchen
Reviewed-by: @limingqi107,@cristoval
Signed-off-by: @cristoval
4 years ago
cbb243a207
print graph output
4 years ago
ba0b4c9e35
!13974 fix conv1x1
...
From: @zhujingxuan
Reviewed-by: @wangchengyuan
Signed-off-by: @wangchengyuan
4 years ago
1d5e903771
!13978 Reduce/Transpose/TensorAdd CPU kernel performance improve！
...
From: @yang_chun
Reviewed-by: @wuxuejian,@c_34
Signed-off-by: @wuxuejian,@c_34
4 years ago
7e1dd06077
!13961 Get correct monad
...
From: @liangzelang
Reviewed-by:
Signed-off-by:
4 years ago
7f92400ebd
!14068 [GraphKernel] fix graph_split and checkcircle
...
From: @lingyunli63
Reviewed-by: @gaoxiong1,@dylangeng
Signed-off-by: @dylangeng
4 years ago
e6709f7573
fix minimum grad bug
4 years ago
defb589076
fix some errors in docs of probability
4 years ago
4dab4521c3
unfold repeated label in labelswitches
4 years ago
c57f106249
Update API document of ScatterNDUpdate
4 years ago
e1d443efe3
tensor-rt library dynamic loadg
4 years ago
f23ce6c7d9
optimize record gpu memory information
4 years ago
41698dbf39
free buffer when program exit
4 years ago
32b19aef59
fix executor last task not clear
4 years ago
f1fb0d9f3a
!13833 Add SparseToDense
...
From: @ZhengQihao3f3f3f
Reviewed-by:
Signed-off-by:
4 years ago
c7b09162c7
!12009 rearch const tensor decompress, copy and dequant
...
From: @hangangqiang
Reviewed-by:
Signed-off-by:
4 years ago
0f365cd017
fix docs
4 years ago
e993ec2efa
!14121 fix sponge ops master
...
From: @jiahongqian
Reviewed-by: @ljl0711,@wang_zi_dong
Signed-off-by: @wang_zi_dong
4 years ago
2a52854da5
!14081 Increase exception capture in checking cuda version
...
From: @gaoyong10
Reviewed-by: @wilfchen,@cristoval
Signed-off-by: @wilfchen
4 years ago
5c6745b3f5
update_case_for_interface_forwardvalueandgrad
4 years ago
5d0490909d
!14097 map a fv cnode if it is not mapped and belong to primal_graph_
...
From: @xychow
Reviewed-by: @ginfung,@zh_qh
Signed-off-by: @zh_qh
4 years ago
b5a25045f8
!14113 incorporate tuple_getitem(switch()) if g1 or g2 is a specific kind of fprop funcgraph
...
From: @xychow
Reviewed-by: @zh_qh,@ginfung
Signed-off-by: @zh_qh
4 years ago
34586542bc
!14087 fix assign execution order problem
...
From: @Margaret_wangrui
Reviewed-by: @zh_qh,@hwhewei
Signed-off-by: @zh_qh
4 years ago
a4b50593c0
!14017 Keep states between hyper map inner ops.
...
From: @zh_qh
Reviewed-by: @zhunaipan,@ginfung,@hwhewei
Signed-off-by: @ginfung
4 years ago
4fa6c4b2e1
!13894 fix RandomChoiceWithMask's random property
...
From: @TFbunny
Reviewed-by:
Signed-off-by:
4 years ago
057db8fedd
fix randomchoicewithmask's random property
4 years ago
f1071fc9ff
!14109 Fixes numpy.take fail in graph mode
...
From: @wangrao124
Reviewed-by: @c_34,@liangchenghui
Signed-off-by: @c_34
4 years ago
5e14596cd8
fix mindrt cpplint warning
4 years ago
aa7fdf53a7
master 03252 sponge ops zoo cu
4 years ago
ac5371b38f
!10774 add gpt2 model zoo
...
From: @ouyang_xx
Reviewed-by:
Signed-off-by:
4 years ago
a8a0956362
!13790 Add fix to IndexAdd GPU
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @liangchenghui
4 years ago
a07d67a441
!14102 modify nn annotation
...
From: @changzherui
Reviewed-by: @kingxian,@zhoufeng54
Signed-off-by: @kingxian
4 years ago
d2ecf71ace
!13693 Slice op only support input_x int32 and float32 at CPU backend
...
From: @wangyanling10
Reviewed-by:
Signed-off-by:
4 years ago
2f00c1dbaa
!14078 oneclick on gate
...
From: @zhaozhenlong
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
4 years ago
33c9fda50a
!13952 fix NoRepeatNGram example
...
From: @yanzhenxiang2020
Reviewed-by: @wuxuejian,@liangchenghui
Signed-off-by: @wuxuejian
4 years ago
2d73a35793
!14056 tensorprint segmentation
...
From: @yepei6
Reviewed-by: @kingxian,@kisnwang
Signed-off-by: @kingxian
4 years ago
d91cf86d2d
Add nn.Conv3d and nn.Conv3dTranspose.
4 years ago
a48785cdcc
!14052 add op atomic clean to clear input addr in launch allreduce
...
From: @lvchangquan
Reviewed-by: @kisnwang,@chujinjin
Signed-off-by: @chujinjin
4 years ago
0e60235107
[MSLITE] Fix the bug of matmul int8 for arm32
4 years ago
0909d69b4e
!14067 fix graph compile time over 3h
...
From: @zyli2020
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
4 years ago
d519b22b76
!13975 fix mindspore numpy doc errors
...
From: @yanglf1121
Reviewed-by: @liangchenghui,@kingxian
Signed-off-by: @liangchenghui,@kingxian
4 years ago
74d500a756
If a fv cnode is not mapped but belong to primal_graph_, then don't propagate this fv to caller, it should mapped instead
4 years ago
cd66730ee4
if g1 or g2 is a fprop funcgraph which bprop funcgraph has env_getitem/env_setitem, then incorporate that tuple_getitem(switch) to eliminate env_setitem
4 years ago
b31c5fab61
add constexpr for unimplemented error
4 years ago
a4db0f862d
!14055 Add clear error log when set device id failed
...
From: @zyli2020
Reviewed-by: @wilfchen,@limingqi107,@wilfchen
Signed-off-by: @wilfchen
4 years ago
b141cb879c
!14061 [lite]fix fill infer bug
...
From: @xu_anyue
Reviewed-by: @hangangqiang,@HilbertDavid
Signed-off-by: @hangangqiang
4 years ago
142cbaf4ce
fix fp16
4 years ago
6c51aa3fdd
modify nn annotation
4 years ago
4b04345393
fix assign execution order problem
4 years ago
a76ea6f828
!14093 update stm32f746 example codes
...
From: @zoloft
Reviewed-by: @wangchengyuan,@zhanghaibo5
Signed-off-by: @wangchengyuan
4 years ago
c1a802627b
!14013 Enable zero dimension using attribute not parameter
...
From: @liangzhibo
Reviewed-by: @zh_qh,@ginfung
Signed-off-by: @zh_qh
4 years ago
c0a6117798
!14057 [MS][LITE] Support encoder fp16
...
From: @cjh9368
Reviewed-by: @hangangqiang,@zhanghaibo5
Signed-off-by: @hangangqiang
4 years ago
2578a6cb65
update stm32f746 example codes
4 years ago
f511830d1d
oneclick ongate
4 years ago
b260430ab1
!14025 [MS][LITE][CPU] transpose bug fix
...
From: @lzkcode
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
4 years ago
114c894be2
!13269 add actor runtime interface
...
From: @limingqi107
Reviewed-by:
Signed-off-by:
4 years ago
e84a8acba9
!13770 fix typo errors and args mismatches
...
From: @dong-li001
Reviewed-by: @zhang_yi2020,@kingxian,@zhunaipan
Signed-off-by: @kingxian
4 years ago
8f003249d1
Increase exception capture in checking cuda version
4 years ago
36797a53a8
[MS][LITE] support encoder fp16
4 years ago
bbd9cb9f69
!14036 up_date mnist_x86 example codes
...
From: @zoloft
Reviewed-by: @wangchengyuan,@hangangqiang
Signed-off-by: @wangchengyuan
4 years ago
6e997ad3fc
!14003 [MD][Profiling] Fix Device ID profiling error
...
From: @xiefangqi
Reviewed-by: @liucunwei,@heleiwang
Signed-off-by: @liucunwei
4 years ago
8832248205
fix abstract null and bug in checkcircle
4 years ago
3f9d9c5b2e
add error log when set device id failed
4 years ago
d2598d73be
fix graph compiling cost time over 3h
4 years ago
c330ff9221
!14047 [MS][LITE][Develop]fix use of bind thread
...
From: @lx0095
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
4 years ago
27f508760b
Add SparseToDense op
4 years ago
a380ee8bf2
!13927 mindspore lite java add ci
...
From: @yeyunpeng2020
Reviewed-by:
Signed-off-by:
4 years ago
06e8c677ee
!13989 Fix float16 support problem and add raises for Batch Dot ops
...
From: @anrui-wang
Reviewed-by: @wuxuejian,@c_34
Signed-off-by: @wuxuejian
4 years ago
a977afe180
!13998 【MS】【LITE】【GPU】fix opencl infershape bug
...
From: @wangdongxu6
Reviewed-by: @zhanghaibo5,@ddwsky
Signed-off-by: @ddwsky
4 years ago
9b17c14896
!13901 unify check fuction
...
From: @lianliguang
Reviewed-by: @kisnwang,@chujinjin
Signed-off-by: @chujinjin
4 years ago
3d2093b666
!14038 [MSLITE][DEVELOP] fix bug of npu convolution
...
From: @yangruoqi713
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhang_xue_tong
4 years ago
16b746f703
!14040 fix make equal less npu-unsupported case
...
From: @zhaozhenlong
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
5926f7c386
!13906 【MD】【BUG】Conventional judgment on the parameter h w not less than zero
...
From: @xulei2020
Reviewed-by:
Signed-off-by:
4 years ago
c792d76c06
fix fill infer bug
4 years ago
ee87860eeb
Reduce/Transpose/TensorAdd CPU kernel performance improve!
4 years ago
2b24a7a9cf
Keep states between hyper map inner ops.
4 years ago
af6d812436
enable zero tensor using attribute
4 years ago
0f28c1aa19
add the force cast to avoid the segmentation
4 years ago
27f7bfef50
!13996 [lite]gelu fusion and fp16 slice
...
From: @xu_anyue
Reviewed-by: @hangangqiang,@HilbertDavid
Signed-off-by: @hangangqiang
4 years ago
0a7df321fe
add op atomic clean to clear input addr in launch allreduce
4 years ago
8629c995c5
!13895 march native coused core dump
...
From: @donghufeng
Reviewed-by:
Signed-off-by:
4 years ago
b9f87ba48a
!14024 [MS_LITE] fix bug decoder fp16
...
From: @YeFeng_24
Reviewed-by: @hangangqiang,@zhanghaibo5
Signed-off-by: @hangangqiang
4 years ago
b7ea5a3ef0
fix bind thread
4 years ago
942b8d7e4e
fix 1x1
4 years ago
f065f41167
[MSLITE][DEVELOP] fix bug of npu convolution
4 years ago
624a504b3b
fix equal 2d less 1d segment fault
4 years ago
d9f4659cfd
change some check type api
4 years ago
946e403df8
!13701 dynamic memory pool support multi-thread
...
From: @limingqi107
Reviewed-by: @cristoval,@kisnwang
Signed-off-by: @kisnwang
4 years ago
dfd368a574
!12757 Change all io_format in master
...
From: @liangzhibo
Reviewed-by: @kingxian
Signed-off-by: @kingxian
4 years ago
26e20de832
up_date mnist_x86 example codes
4 years ago
4b329090b6
!13938 library size optimization for 310
...
From: @zhoufeng54
Reviewed-by: @xu-yfei,@kisnwang
Signed-off-by: @xu-yfei
4 years ago
a55cc5d651
add java ci
4 years ago
d4f49b3a10
fix opencl infershape bug
4 years ago
ca38cff71d
fix model buf may double free
4 years ago
3b6dabc05b
transpose bug fix
4 years ago
46745e2a21
fix_fp16_door
4 years ago
18e98c6a0b
!13720 【GraphKernel】Add context graph_kernel_flags
...
From: @dayschan
Reviewed-by: @gaoxiong1
Signed-off-by:
4 years ago
49e39c3bdf
!13992 update the documentation of StridedSlice, Gather, GatherNd, etc. operators.
...
From: @wangshuide2020
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
fc360dc6ff
!13911 [MD] Fix an TestStepParallel core dump
...
From: @xiefangqi
Reviewed-by: @pandoublefeng,@liucunwei
Signed-off-by: @pandoublefeng,@liucunwei
4 years ago
4122d1dcce
fix code format bug
4 years ago
2908702204
layernorm fusion and gelu fusion
4 years ago
b808876234
!13987 [MS][LITE]fix a bug of the fc op in the fp16 subgraph and add models to the entrance guard
...
From: @probiotics_53
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhanghaibo5
4 years ago
4f0ad9398e
!13969 [MSLITE] Fix bug of pad operator.
...
From: @wang_shaocong
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
7647c06bfc
!13839 [MS][LITE][Develop]add GetOutputsByNodeName
...
From: @lx0095
Reviewed-by: @zhang_xue_tong
Signed-off-by:
4 years ago
69dfcb3936
!13985 fix weight quant may print error log
...
From: @xutianchun
Reviewed-by: @zhang_xue_tong,@zhanghaibo5
Signed-off-by: @zhanghaibo5
4 years ago
f2bae5f93c
!13946 [MS_LITE] fix decoder fp16
...
From: @YeFeng_24
Reviewed-by: @hangangqiang
Signed-off-by:
4 years ago
73816003c1
!13965 fix warning error of generated code
...
From: @yangjie159
Reviewed-by: @wangchengyuan,@zhang_xue_tong
Signed-off-by: @wangchengyuan
4 years ago
9b23952fc2
!13867 add pad and strided slice fusion npu
...
From: @zhaozhenlong
Reviewed-by:
Signed-off-by:
4 years ago
a78c37ebcb
!13981 modify release.md, add description for Depend/ControlDepend.
...
From: @huangbingjian
Reviewed-by: @hwhewei,@ginfung
Signed-off-by: @ginfung
4 years ago
fd75d4866f
modify release.md, add description for Depend/ControlDepend
4 years ago
9023ba2b56
!13745 Enable zero dimension when it comes out of operator.
...
From: @liangzhibo
Reviewed-by: @ginfung,@zh_qh,@ginfung
Signed-off-by: @zh_qh
4 years ago
dbe4e7ff82
md fix gpu profiling device id issue
4 years ago
be0e2f18e6
!14000 update documentation of ExponentialDecayLR, PolynomialDeca, etc.
...
From: @mind-lh
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
fcce705bcb
!13818 revert nn.BatchNorm3d.
...
From: @liu_xiao_93
Reviewed-by: @liangchenghui
Signed-off-by: @liangchenghui
4 years ago
2e7f9efebd
!13811 fix sponge ops
...
From: @jiahongqian
Reviewed-by: @wang_zi_dong,@ljl0711,@wang_zi_dong
Signed-off-by: @ljl0711
4 years ago
7aeeed7e4d
fix a bug of the fc op in the fp16 subgraph and add models to the entrance guard
4 years ago
3ca1b4fc81
CUDNN STATUS NOT SUPPORTED
4 years ago
13dc35973a
fix_decoder_3-master
4 years ago
be4ffe94ce
fix code docs
4 years ago
32f35b1055
npu pad stridedslice fusion for nasnet
4 years ago
05df8a9b26
XXXLR not support cpu: ExponentialDecayLR, PolynomialDecayLR, etc.
4 years ago
1e6e4f6027
[MSLITE] Fix bug of the padding parameter of onnx pad operator.
4 years ago
a47bf2b499
update the documentation of StridedSlice, Gather, GatherNd, etc. operators.
4 years ago
1307f203c5
fix weight quant may print error
4 years ago
761a2e2127
Bug fix for ISSUE #I3CN9Q
4 years ago
cc99846700
!13935 fix mobilenetv2 readme.md
...
From: @zoloft
Reviewed-by: @wangchengyuan,@wangchengyuan,@zhanghaibo5
Signed-off-by: @wangchengyuan
4 years ago
f94d660626
march native cause whl core dump
4 years ago
a54d6ca8ff
fix warning error of generated code
4 years ago
17384bfbba
!13898 [MS][LITE][CPU] fix transpose bug
...
From: @lzkcode
Reviewed-by: @hangangqiang,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
1e4a086ffd
[Auto-monad] get correct monad
4 years ago
e4be751742
!13845 Update API documentation of MatrixDiag, ScatterND and NMSWithMask
...
From: @yuyiyang_3418
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
923dd4b77a
enable zero dimension tensor when input is data
4 years ago
a5fa2f5549
fix mobilenetv2/README.md
4 years ago
d926e8e4f4
!12450 python3.9 update
...
From: @yepei6
Reviewed-by:
Signed-off-by:
4 years ago
b1b99b97d7
!13750 Add check of ResizePreserveARWithFiller operation
...
From: @shenwei41
Reviewed-by:
Signed-off-by:
4 years ago
ec1e44595a
!13823 [MD] fix bug in closing multiprocessing pool
...
From: @liyong126
Reviewed-by: @heleiwang,@liucunwei
Signed-off-by: @liucunwei
4 years ago
e4cca000cc
Update API documentation of MatrixDiag, ScatterND and NMSWithMask
4 years ago
da7e3ff55c
fix NoRepeatNGram example
4 years ago
95657ef48c
!13931 features, add model implement and support parallel
...
From: @yangjie159
Reviewed-by: @wangchengyuan,@wangchengyuan,@zhang_xue_tong
Signed-off-by: @wangchengyuan
4 years ago
baed02fd5f
!13948 Fix error words for api comments.
...
From: @zhang_yi2020
Reviewed-by: @gemini524,@liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
2dc88b2cc0
!13884 Add RANK_TABLE_FILE for PyNative
...
From: @jojobugfree
Reviewed-by: @kisnwang,@chujinjin
Signed-off-by: @chujinjin
4 years ago
980769d404
fix error words for api comments.
4 years ago
c1001e9172
!13848 Fix a bug by revert GetCNodeTarget() for Load
...
From: @hwhewei
Reviewed-by: @ginfung,@zh_qh
Signed-off-by: @zh_qh
4 years ago
fb64e14265
fix slice op bug
4 years ago
f17462859c
!13860 [MS][LITE]multi-inputs models support fp16 mode and add models to the entrance guard
...
From: @probiotics_53
Reviewed-by: @zhanghaibo5,@hangangqiang
Signed-off-by: @zhanghaibo5
4 years ago
1200ae2f27
features, add model implement and support parallel
4 years ago
84c1705319
add GetOutputsByNodeName
4 years ago
11ee3b1624
add context graph_kernel_flags
...
used the flag "opt_level" to control GraphKernel,
0 means disabled while non-zero value means enabled.
the default value is controlled by context "enable_graph_kernel",
but if it's also set in "graph_kernel_flags", then the flag will prevail.
supported the whitelist and blacklist operators for GraphKernelExpander.
"enable_expand_ops", "enable_expand_ops_only", "disable_expand_ops".
4 years ago
56f3776d7f
add actor runtime interface
4 years ago
39623cf2dd
multi-inputs models support fp16 mode and add models to the entrance guard
4 years ago
3f187e22e5
Fix an ut cpp core issue
4 years ago
b1c86b6a22
!13857 check whether communication unit has been inited
...
From: @yao_yf
Reviewed-by: @kisnwang,@stsuteng
Signed-off-by: @stsuteng
4 years ago
9ac13fb8e2
!13885 fix quant _partial_init example
...
From: @yuchaojie
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @zhoufeng54
4 years ago
159071d438
!13882 add README for mobilenetv2
...
From: @zoloft
Reviewed-by: @wangchengyuan,@zhujingxuan,@hangangqiang,@zhanghaibo5
Signed-off-by: @wangchengyuan
4 years ago
e4c59a1c8f
library size optimization for 310
...
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
4 years ago
0f900c8d53
power bug fix
4 years ago
1732ed34b6
!13877 remove ControlDepend from ternarybert
...
From: @huangbingjian
Reviewed-by: @ginfung,@hwhewei
Signed-off-by: @hwhewei
4 years ago
e0b1afcfe5
!13819 [MS][LITE][CPU]fix bug of floormod
...
From: @fuzhiye
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
d3265f0591
Add check of ResizePreserveARWithFiller operation
4 years ago
7e647bdc6f
add fix to IndexAdd
4 years ago
4f16b2542e
update python3.9
4 years ago
a59f1814c1
!13835 fixed memcpy error
...
From: @anancds
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
4 years ago
444ff97206
!13505 typo fix
...
From: @wudenggang
Reviewed-by: @ouwenchang,@Hanshize,@kingxian
Signed-off-by: @kingxian
4 years ago
34fecc0697
!13864 update applyAdagrad
...
From: @zyx5256
Reviewed-by: @wuxuejian,@liangchenghui
Signed-off-by: @wuxuejian
4 years ago
f2830073b5
fix quant _partial_init example
4 years ago
83c2817fce
!13840 fix Vector && script
...
From: @zhujingxuan
Reviewed-by: @wangchengyuan,@jpc_chenjianping
Signed-off-by: @wangchengyuan
4 years ago
0c76dcb721
add RANK_TABLE_FILE for PyNative Hccl
4 years ago
b88195a715
add README for mobilenetv2
4 years ago
69af6643d3
!13703 [lite]matmul and add fusion
...
From: @xu_anyue
Reviewed-by:
Signed-off-by:
4 years ago
4ef7251e7f
remove ControlDepend from ternarybert
4 years ago
723bbac438
revert nn.BatchNorm3d.
4 years ago
ae3127f12f
!13808 [auto_monad]Remove uselesee updatestate pattern and fix loss and J order.
...
From: @linqingke
Reviewed-by: @hwhewei,@zh_qh
Signed-off-by: @zh_qh
4 years ago
6cd1a0b484
!13760 fix bug of quant export
...
From: @chengxb7532
Reviewed-by:
Signed-off-by:
4 years ago
990aaba139
!13604 add compile info for unpack return value
...
From: @liubuyu
Reviewed-by: @zhoufeng54,@jjfeing,@kisnwang
Signed-off-by: @kisnwang
4 years ago
a12ae4286a
!13821 skip empty graph in profiling
...
From: @jojobugfree
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @zhoufeng54
4 years ago
d8c66aee4a
!13715 fix numpy native cpu ci error
...
From: @yanglf1121
Reviewed-by: @guoqi1024,@liangchenghui
Signed-off-by: @liangchenghui
4 years ago
0cf0663a50
!13837 fix gthread_num
...
From: @zoloft
Reviewed-by: @wangchengyuan
Signed-off-by: @wangchengyuan
4 years ago
57f7bf8480
!13791 reduce cast number in syncbn
...
From: @yuchaojie
Reviewed-by: @kisnwang,@zhoufeng54
Signed-off-by: @zhoufeng54
4 years ago
87c97ecb26
!13810 fix error of context field not defined
...
From: @wangnan39
Reviewed-by: @kingxian,@zh_qh
Signed-off-by: @kingxian,@zh_qh
4 years ago
a11287c332
adagrad: support ouput on gpu
4 years ago
a767b1a3e2
fix Vector && script
4 years ago
f5763bdebb
check whether do init in distributed scene
4 years ago
684398e755
fix gthread_num
4 years ago
c66b4e4994
!13729 fix TrainOneStepWithLossScaleCell execorder
...
From: @youui
Reviewed-by: @kisnwang,@guoqi1024
Signed-off-by: @guoqi1024
4 years ago
c4e339674b
!13806 add air restriction note
...
From: @caozhou_huawei
Reviewed-by: @kingxian,@zh_qh
Signed-off-by: @kingxian
4 years ago
d668baf979
matmul and add fusion
4 years ago
0769193185
Fix a bug by revert GetCNodeTarget() for Load
4 years ago
a45f03eebf
!13490 fix bugs, and add dump tensor in benchmark
...
From: @yangjie159
Reviewed-by: @wangchengyuan,@zhanghaibo5
Signed-off-by: @wangchengyuan
4 years ago
3bd84e3705
!13782 update example and run script
...
From: @zhujingxuan
Reviewed-by: @wangchengyuan,@wangchengyuan,@zhanghaibo5
Signed-off-by: @wangchengyuan,@wangchengyuan
4 years ago
902b4de2f8
!13714 update documentation of Index_add operator,input of Slice,example of Unstack.
...
From: @mind-lh
Reviewed-by:
Signed-off-by:
4 years ago
20d881c5be
fixed memcpy error
4 years ago
9629977242
fix a bug that cuda error show wrong name in log
4 years ago
96bfe123a2
1.update the documentation of IndexAdd operator
...
2.update the input of Slice
3.update the example of Unstack.
4 years ago
bfcf4562ad
!13582 【MD】【BUG】fix MD lite train error
...
From: @xulei2020
Reviewed-by:
Signed-off-by:
4 years ago
85b3389caa
fix bug of quant export
4 years ago
dc03f62f4b
fix bug of using pool
4 years ago
0fa621ea0d
!13666 [MS][LITE]append some models to the entrance guard
...
From: @probiotics_53
Reviewed-by:
Signed-off-by:
4 years ago
fa64151dd8
!13781 Ingnore transpose with nop_op attr in pynative infer
...
From: @HulkTang
Reviewed-by: @zhoufeng54,@chujinjin
Signed-off-by: @chujinjin
4 years ago
acfa8f066a
!13699 [MS][LITE][Develop]fix write for big endian devices
...
From: @lx0095
Reviewed-by:
Signed-off-by:
4 years ago
ad4dedb1db
!13800 clear memory after run op in CPU
...
From: @simson_wu
Reviewed-by: @chujinjin,@zhoufeng54
Signed-off-by: @chujinjin
4 years ago
af8eebce84
!13804 fix micro thread_pool
...
From: @zoloft
Reviewed-by: @wangchengyuan,@hangangqiang
Signed-off-by: @wangchengyuan
4 years ago
1e2d538b37
skip empty graph in profiling
4 years ago
b2a6e85a7a
fix bug of floorMod
4 years ago
619a6e8c96
jiahong sponge ops 03232
4 years ago
baa35ab3f9
!13609 add kwargs for vgg16
...
From: @caojian05
Reviewed-by: @wuxuejian,@oacjiewen
Signed-off-by: @wuxuejian
4 years ago
bf7c1049dc
!13742 Add ENV_TRACE_LABEL_WITH_UNIQUE_ID for enabling unique id print.
...
From: @zh_qh
Reviewed-by:
Signed-off-by:
4 years ago
dd2cd6d251
fix error of context field not defined
4 years ago
fb69f3e509
!13794 [MS][LITE][CPU]fix bug of reduce
...
From: @fuzhiye
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
db385e9b05
remove uselesee updatestate pattern and fix loss and J order
4 years ago
ada56f3c42
fix micro thread_pool
4 years ago
c27ecbcdae
add air restriction note
4 years ago
12a29ce040
!13751 [MSLITE][Develop] fix bug of arm cpu op gather while running in multi-threaded mode
...
From: @yangruoqi713
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
aa7b49fe32
clear memory after run op in CPU
4 years ago
e769ee95ce
reduce cast number in syncbn
4 years ago
287988553e
fix bug of reduce
4 years ago
ab5d77f943
!13773 fixed core dump
...
From: @anancds
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
4 years ago
5d1a340fb4
!13780 add mnist stm32f746 example
...
From: @zoloft
Reviewed-by: @wangchengyuan,@hangangqiang
Signed-off-by: @wangchengyuan
4 years ago
mindspore-ci-bot
zengxianglong
tronzhang
liuxiao93
zhujingxuan
z00512249
LianLiguang
wang_shaocong
ms_yan
yepei6
lingyunli63
buxue
lixian
liuhe
dingpeifei
pkuliuliu
looop5
Jiaqi
zengzitao
mengyuanli
yeyunpeng2020
yangruoqi713
Ziyan
linqingke
dayschan
wilfChen
limingqi107
kswang
caifubi
zhaozhenlong
zhangxinfeng3
xiefangqi
wangnan39@huawei.com
liangzelang
shenwei41
chenfei
zhengjun10
yuyiyang_3418
louei5
chenjianping
huangmengxi
lvliang
TFBunny
wangcong
q00596439
zhanyuan
zhousiyi
wangrao
hangangqiang
changzherui
Margaret_wangrui
gaoyong10
cjh9368
lizhenyu
zhengqihao
xuanyue
yang_chun
Zhang Qinghua
l00591931
lvchangquan
wangdongxu
xutianchun
lzk
yefeng
xulei2020
huangbingjian
wangshuide2020
w00535372
donghufeng
yangjie159
yanzhenxiang2020
zhangyi
wangyanling
zhoufeng
yuchaojie
zhuyuxiao
yao_yf
He Wei
chendongsheng
hanhuifeng2020
chengxianbin
liyong
fuzhiye
caozhou
simson