7b5eaac570
!13223 zero copy
...
From: @zhupuxu
Reviewed-by:
Signed-off-by:
4 years ago
f6677628c0
add StackPush and StackPop for aicpu
4 years ago
5a6bb251b0
!12724 add Dropout2D and rename Dropout3d to Dropout3D
...
From: @yanzhenxiang2020
Reviewed-by:
Signed-off-by:
4 years ago
3cc573b5b8
zero copy
...
Signed-off-by: zhupuxu <zhupuxu@huawei.com>
4 years ago
621b796ed3
fix error when train twice in pynative and feed model
4 years ago
423026c1ac
!13226 update graphengine 0311
...
From: @shenwei41
Reviewed-by:
Signed-off-by:
4 years ago
c0f966ddef
update graphengine 0311
4 years ago
4d35303265
support string in GPU print
4 years ago
e1a95b1152
!12714 rebuild getitem
...
From: @yepei6
Reviewed-by:
Signed-off-by:
4 years ago
762c63af6f
Add error message
4 years ago
063607de90
!13015 update testcases of dump
...
From: @rainyhorse
Reviewed-by: @xsmq,@zhoufeng54
Signed-off-by: @zhoufeng54
4 years ago
3f397503c0
add dynamic shape testcases for Sub
4 years ago
2303abd58c
!12693 Setting of dynamic_batch_size supported
...
From: @zhupuxu
Reviewed-by:
Signed-off-by:
4 years ago
5ca25451ee
dynamic_batch
...
Signed-off-by: zhupuxu <zhupuxu@huawei.com>
4 years ago
70024d3ab1
!12189 Add CPU Pad op
...
From: @wanyiming
Reviewed-by:
Signed-off-by:
4 years ago
d2432a962a
update the convert tool path
4 years ago
723844a06f
!12999 Dvpp input validation
...
From: @lizhenglong1992
Reviewed-by: @liucunwei
Signed-off-by: @liucunwei
4 years ago
57373ed30a
merge getitem the by otehr types to by tuple(success)
4 years ago
750d7e6e2a
!12793 [auto-monad] Refactor ascend_auto_monad
...
From: @hwhewei
Reviewed-by:
Signed-off-by:
4 years ago
2474a0237f
Fix dvpp input validation and Aipp config generation issue
4 years ago
802e756c9b
!12897 Add float64 support to reducemax grad gpu op
...
From: @peilin-wang
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
8b733dccaa
!12922 biasadd report error when input 5d with nchw format at cpu
...
From: @wangyanling10
Reviewed-by:
Signed-off-by:
4 years ago
dbc0ad13db
cpu_pad_op
4 years ago
b8fc22d551
!12976 New gpu op TensorScatterUpdate
...
From: @peilin-wang
Reviewed-by: @liangchenghui,@wuxuejian
Signed-off-by: @liangchenghui
4 years ago
e6693ea89d
biasadd operator report error when input is not 2d or 4d
4 years ago
dd72f44b27
tensor_scatter_update new op quick initial commit
...
fix ci
fix ci
fix ci
fix ci
4 years ago
d8ec8f6e95
!12782 support load mindir and run
...
From: @wangnan39
Reviewed-by:
Signed-off-by:
4 years ago
891fd7df92
[auto-monad] Refactor ascend_auto_monad
...
1. Remove output parameter pool for ascend control flow;
2. Remove duplicate code for Switch and SwitchLayer;
3. Add 'return' attribute to label_goto that used for return;
4. Disable tail call optimize for graphs with 'recursive' flag.
4 years ago
9ed9d950e2
!11446 fix argmax op for cpu and gpu
...
From: @xcnick
Reviewed-by:
Signed-off-by:
4 years ago
0681c2597b
develop mindir load and run
4 years ago
48d4cca512
!12949 Add TopoSort Rhs First attribute for special CNode, such as Depend CNode with isolated nodes.
...
From: @zh_qh
Reviewed-by: @hwhewei,@zhunaipan
Signed-off-by: @zhunaipan
4 years ago
df866f7248
Add TopoSort Rhs First attribute for special CNode, such as Depend CNode with isolated nodes.
4 years ago
df98fa0233
!12842 Add grad op for gpu dynamic shape testing op
...
From: @peilin-wang
Reviewed-by:
Signed-off-by:
4 years ago
d285692217
!12852 Change maketuple in coreops
...
From: @liangzhibo
Reviewed-by:
Signed-off-by:
4 years ago
26f6fea675
debug_for_new_interface_forwardvalueandgrad
4 years ago
680324f225
Change make tuple in core.ops
4 years ago
8432bdffbf
!12880 move operator primitive of Centralization to _inner_ops
...
From: @david-he91
Reviewed-by: @liangchenghui,@guoqi1024
Signed-off-by: @liangchenghui
4 years ago
783c57c209
add float64 support to reducemax grad
...
fix ci
4 years ago
56e8aa9f7e
change index_add op input_x to type Parameter
4 years ago
d65a5affba
fix cpu/gpu argmax op
4 years ago
d8382f4d57
move operator primitive of Centralization to _inner_ops
4 years ago
8b8c59f01e
Optimize the compile performance in Parser, FG, Manager and Renormalize:
...
---
Remove the routine of handling isolated nodes in Renormalize.
Add isolated nodes from Parser&Resolver.
Modify isolated nodes handling in FG&Manager.
Optimize the renormalize routines.
Other optimizations.
4 years ago
bb303647c7
!12799 Fix typos in head descriptions about license.
...
From: @c_34
Reviewed-by: @wuxuejian
Signed-off-by: @wuxuejian
4 years ago
9aa0753588
!12840 open 8p testcases
...
From: @zhoufeng54
Reviewed-by:
Signed-off-by:
4 years ago
7ff2b3b499
!12781 fix bug of amp bn cast
...
From: @jojobugfree
Reviewed-by:
Signed-off-by:
4 years ago
541643e16b
!12777 fix ci pynative case
...
From: @jojobugfree
Reviewed-by: @chujinjin,@jjfeing
Signed-off-by: @chujinjin
4 years ago
347575a38f
add grad for GpuDynamicShape testing op
...
add tests
removed some unused imports
fix ci
4 years ago
98775f29dd
open 8p testcase
...
Signed-off-by: zhoufeng <zhoufeng54@huawei.com>
4 years ago
ae1e6bf176
fix ci pynative case
4 years ago
4b7e76b4fb
!12796 AIpp config file generator bug fix
...
From: @lizhenglong1992
Reviewed-by: @liucunwei,@heleiwang
Signed-off-by: @liucunwei
4 years ago
82c5486714
fix typos in lincense
4 years ago
99c17c43d8
!12746 Fix ApplyAdagradCpuKernel output
...
From: @yang_chun
Reviewed-by: @wuxuejian,@c_34
Signed-off-by: @wuxuejian
4 years ago
07dd752e1e
Fix aipp config file generator bug
4 years ago
a6959c2a13
fix bn cast bug
4 years ago
6459229fff
adapt some 3d ops.
4 years ago
db0a6f1e19
replace ps-lite
4 years ago
d61f646431
!12743 Add ERROR and WARNING information for execute class
...
From: @lizhenglong1992
Reviewed-by: @heleiwang,@pandoublefeng
Signed-off-by: @pandoublefeng
4 years ago
580d33e90e
Fix: ApplyAdagrad output.
4 years ago
3a319a97e9
move matmul from numpy to ops
4 years ago
5cf805d46a
Fix some issue of ERROR information in log
4 years ago
95dbfe0636
add Dropout2D and rename Dropout3d to Dropout3D
4 years ago
c33d767314
!12672 Aipp config file generation
...
From: @lizhenglong1992
Reviewed-by:
Signed-off-by:
4 years ago
692d158f5c
!12656 numpy-native repeat add checking for repeats shape
...
From: @jachua
Reviewed-by:
Signed-off-by:
4 years ago
22e5ed13d6
Aipp config file generation
4 years ago
fd3cec2e9f
!12627 update graphengine_0225
...
From: @shenwei41
Reviewed-by:
Signed-off-by:
4 years ago
d10dc28eb2
update graphengine_0225
4 years ago
03935de4bf
!12630 make CPU op ResizeNearestNeighbor support float64 and Int64
...
From: @wanyiming
Reviewed-by:
Signed-off-by:
4 years ago
d8433e9f5c
!12605 Fix GPU ReduceMax/Sum float64 type failure
...
From: @TFbunny
Reviewed-by: @tom__chen
Signed-off-by:
4 years ago
10692e9cd7
!12663 [auto-monad] Fix backend control flow bug found by igamma test
...
From: @hwhewei
Reviewed-by: @zh_qh,@ginfung
Signed-off-by: @zh_qh
4 years ago
ca6ffac0e8
add_int64_float64
4 years ago
76c8e10c4a
bug fix
4 years ago
09eee17406
!12591 Add sinh, cosh, asinh, acosh, atanh, atan2, asinhgrad, acoshgrad for CPU
...
From: @wangrao124
Reviewed-by:
Signed-off-by:
4 years ago
c837fb25a2
[auto-monad] Fix backend control flow bug found by igamma test
4 years ago
f269f3fc8d
!12231 Support ad for scalar inputs
...
From: @ginfung
Reviewed-by:
Signed-off-by:
4 years ago
ec644a5291
add fix to float64 support in reduce ops
4 years ago
03b1aeecdb
FusedBatchNormEx dynamic shape support
4 years ago
ab17c49eba
add sinh, cosh, asinh, acosh, atanh, atan2, asinhgrad, acoshgrad for cpu
4 years ago
41189781f3
support scalar input for cell
4 years ago
88484a4c83
!12556 add dynamic shape testcase to AllReduce GPU
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
a38c996c9c
!12537 Dvpp API merge
...
From: @lizhenglong1992
Reviewed-by:
Signed-off-by:
4 years ago
09f1003617
!12504 Add ascend profiling st
...
From: @jojobugfree
Reviewed-by: @zhoufeng54,@chujinjin
Signed-off-by:
4 years ago
581f1e82d4
API merge with new API class and new IR class
4 years ago
50bf5033f2
!12400 Add float64 support to ZerosLike, remove duplicate ZerosLike gpu kernel
...
From: @peilin-wang
Reviewed-by:
Signed-off-by:
4 years ago
9f48ab87c0
!12524 Add DType dynamic shape testcases for GPU
...
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
4 years ago
f260e11c0f
adding dynamic shape testcase to all reduce
4 years ago
f1f0ad41ff
update tests/st/networks/models/resnet50/test_resnet50_imagenet.py.
4 years ago
afcbf45019
!12506 Add testcase related to auto-monad
...
From: @huangbingjian
Reviewed-by:
Signed-off-by:
4 years ago
1d640e50c9
!12230 add dynamic st test
...
From: @hwjiaorui
Reviewed-by:
Signed-off-by:
4 years ago
42f995e0dd
!12513 [auto-monad] Revert "Change backend execution order sorting policy"
...
From: @hwhewei
Reviewed-by: @zh_qh,@ginfung
Signed-off-by: @zh_qh
4 years ago
57f6c17933
!12438 Add backward op for IndexAdd GPU
...
From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by: @robingrosman
4 years ago
cbc75810a9
add testcases to DType dynamic shape
4 years ago
ba9ae8097f
removed zeroslike from math/unary_op_gpu_kernel and added float64 support
...
fix ci
fix ci
4 years ago
e1462c8684
add indexaddgrad ops support and testcases
4 years ago
cada22e7bf
!12354 Add float64 support to ReduceMax and ReduceSum GPU
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
3c44e731a2
[auto-monad] Revert "Change backend execution order sorting policy"
...
This reverts commit 141c39b71c
4 years ago
a158b2f2fa
add testcase related to auto-monad
4 years ago
0beaa37313
add ascend profiling st
4 years ago
ceaa1603c8
trim down tests & add type checking for meshgrid
4 years ago
90a56c9c23
!12384 Add Sin, Cos, Tan, Atan, AtanGrad for CPU
...
From: @wangrao124
Reviewed-by: @wuxuejian,@kisnwang
Signed-off-by: @wuxuejian
4 years ago
3805f0dfeb
!12355 dataset: Create C++ API classes for vision transform ops
...
From: @cathwong
Reviewed-by:
Signed-off-by:
4 years ago
3fc204e3bc
dataset: Create C++ API classes for vision transform ops
...
Review rework
Add overloaded API Map function
Overload Execute, update st test_de.cc and enhance ut execute_test.cc;
More TensorTransform support for BoundingBoxAugment and UniformAug
Execute support for auto case; UT updates
Rebase updates, including fix CI Ascend compilations
test_de.cc - comment out dvpp tests
Rebase updates
Added back testcase and constructor
Added back constructor
added missing macro
Fixed Macro
removed decode
4 years ago
4f179898aa
!12420 fix resnet thor smoke threshold
...
From: @wangmin0104
Reviewed-by: @wang_zi_dong,@jjfeing
Signed-off-by: @wang_zi_dong
4 years ago
0a2525a522
add sin, cos, tan, atan for cpu
4 years ago
4b805b87ee
!12386 issue #I37W2N & #I37W2L
...
From: @jachua
Reviewed-by: @liangchenghui,@liangchenghui
Signed-off-by: @liangchenghui
4 years ago
52eae8d4ad
!12329 GPU index_add forward op
...
From: @tom__chen
Reviewed-by: @liangchenghui,@robingrosman
Signed-off-by: @liangchenghui
4 years ago
b94bba1def
index_add gpu op
4 years ago
feb07198e7
!12399 Add type support to Squeeze gpu op
...
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
7d97d1300e
!12370 Add type support to CumSum gpu op
...
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
3981b6da0c
add dynamic st test
4 years ago
2e0d1b9940
bug fix
4 years ago
172e918359
fix resnet thor smoke threshold
4 years ago
f2dce335f2
Make the resource manager independent, new base class DeviceResource and new derived class AscendResource
4 years ago
aa71118a99
!12281 fix exec order bug about monad and add test_case in ci
...
From: @zengzitao
Reviewed-by:
Signed-off-by:
4 years ago
191b3f0c8c
!12368 Add type support to Split gpu op
...
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
4821caa143
add uint16, uint32, int8, int64, float64 type support to gpu Squeeze op
...
change to test in graph mode
4 years ago
7443ea4b5b
!12369 Add type support to tile op
...
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
f021af8096
!12367 Add float64 support to MatMul ops in GPU
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
4b27c3206d
!12362 Add float64 support to Abs, Neg, Sqrt, Sin, Cos
...
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
4 years ago
07cb2f3e83
!12283 add new inner operator centralizaiton
...
From: @david-he91
Reviewed-by:
Signed-off-by:
4 years ago
14868354b9
!11649 Add-new-interface-'ForwardValueAndGrad'-to-merge-forward-run-and-grad-operation
...
From: @joylvliang
Reviewed-by:
Signed-off-by:
4 years ago
176afda078
!12307 Add result check for cases in test_cont_grad.py
...
From: @liangzelang
Reviewed-by: @jjfeing,@zhoufeng54
Signed-off-by: @zhoufeng54
4 years ago
dd36171976
add-new-interface-forward-value-and-grad
4 years ago
94054172ce
!12341 fix st error for gpu type support
...
From: @yanglf1121
Reviewed-by: @liangchenghui,@kingxian
Signed-off-by: @liangchenghui,@kingxian
4 years ago
ef3507e973
fix exec order bug about monad
4 years ago
048eba4460
!12359 add float64 support to assign GPU
...
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
4 years ago
6c67bf7cb6
!12223 update testcases of dump
...
From: @rainyhorse
Reviewed-by: @lilongfei15,@kingxian
Signed-off-by: @kingxian
4 years ago
5f7fc5a0b1
initial commit
...
fix ci
4 years ago
1655a88b64
initial commit
4 years ago
2252e440a5
initial commit
...
fix ci
4 years ago
073fc6fa93
initial commit
...
fix ci
4 years ago
4c18e0894e
add float64 support to matmul ops
4 years ago
ce26d2e987
intitial commit
...
fix ci
fix ci
4 years ago
e1ed3b9c15
add float64 support to assign GPU
4 years ago
c6baca71e5
add float64 support to ReduceMax and ReduceSum
4 years ago
8d936a6589
!12352 Add float64 support to Gather GPU
...
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
4 years ago
0ee8ef736d
!12351 Add float64 support to GatherNd GPU
...
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
4 years ago
8fabb26412
!12350 add float64 support to concat GPU
...
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
4 years ago
6f9cfb809e
add float64 support to GatherNd
4 years ago
6fad165e43
add float64 support to gather gpu
4 years ago
ebb7f2e771
add float64 support to concat GPU
4 years ago
d249ededf6
!12347 add float64 support to slice ops in GPU
...
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @tom__chen
4 years ago
ed5f9cb1f2
!11802 Optimize GPU kernels for fasterrcnn
...
From: @robingrosman
Reviewed-by:
Signed-off-by:
4 years ago
799e51cff0
Add float64 support to slice ops
4 years ago
f7dc5e2649
fix gpu_op type error
4 years ago
0660140708
add new inner operator centralizaiton
4 years ago
e2d3495925
Dvpp refactor and 6 Dvpp operators
4 years ago
5d5225f2ee
optimize fasterrcnn on gpu. fix randomchoicewithmask, topk kernels
4 years ago
adf934c567
!12152 Rewrite sequence_mask as a composite op
...
From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by:
4 years ago
1239a4a848
!12314 [auto-monad] Change backend execution order sorting policy
...
From: @hwhewei
Reviewed-by: @ginfung,@zh_qh
Signed-off-by: @zh_qh
4 years ago
141c39b71c
[auto-monad] Change backend execution order sorting policy
...
Change backend execution order sorting policy from DFS to BFS
in KernelGraph::SetExecOrderByDefault().
4 years ago
mindspore-ci-bot
yanzhenxiang2020
zhupuxu
liyong
shenwei41
TFBunny
Zhenglong Li
rainyhorse
yepei6
wanyiming
wangyanling
Peilin Wang
He Wei
wangnan39@huawei.com
Zhang Qinghua
lvliang
l00591931
tom__chen
xcnick
hedongdong
zhoufeng
caifubi
chenhaozhe
liuxiao93
chendongsheng
yang_chun
huangmengxi
wangrao
yujianfeng
wangmin0104
huangbingjian
Cathy Wong
hwjiaorui
zengzitao
yanglf1121
RobinGrosman