682a8926bd
!13176 fix minddata transform issue
...
From: @luoyang42
Reviewed-by:
Signed-off-by:
4 years ago
a1c3f55aca
!13197 support item negative index bprop
...
From: @zhangbuxue
Reviewed-by: @ginfung
Signed-off-by:
4 years ago
50ee325b96
support item negative index bprop
4 years ago
74d7bb41a6
fix minddata transform issue
4 years ago
1e33df94a5
!13142 support name or attribute ast.Expr
...
From: @zhangbuxue
Reviewed-by:
Signed-off-by:
4 years ago
f7040ca134
!13198 Fix get_dataset_size error with PKSampler
...
From: @ziruiwu
Reviewed-by:
Signed-off-by:
4 years ago
1bb789278d
!13200 Fix CI warnings and test team issues
...
From: @alexyuyue
Reviewed-by: @robingrosman,@nsyca
Signed-off-by: @robingrosman
4 years ago
2692e3cc3d
fix pk sampler's get_dataset error due to num_class unavaiable at pre-runtime
4 years ago
236c874e64
!12862 Increase default level of parallelism in minddata pipeline
...
From: @robingrosman
Reviewed-by:
Signed-off-by:
4 years ago
2457f52596
Fix CI warning and test team issues
...
Signed-off-by: alex-yuyue <yue.yu1@huawei.com>
4 years ago
2551f5ea7e
support name or attribute ast.Expr
4 years ago
e1a95b1152
!12714 rebuild getitem
...
From: @yepei6
Reviewed-by:
Signed-off-by:
4 years ago
a0e75b845a
Increase default level of parallelism in mindddata pipeline to 8.
...
Adjust memory usage to not increase as parallelism increases. it will
stay at same level it would be with 4 parallelism
4 years ago
54c37bcd61
!12947 Add MaxPool3D,MaxPool3DGrad,MaxPool3DGradGrad ops for Ascend.
...
From: @liu_xiao_93
Reviewed-by: @liangchenghui
Signed-off-by: @liangchenghui
4 years ago
35f6ba9011
Add MaxPool3D,MaxPool3DGrad,MaxPool3DGradGrad ops for Ascend.
4 years ago
3c7463bf2c
!12943 adding fix to sample script
...
From: @ezphlow
Reviewed-by:
Signed-off-by:
4 years ago
ed373f61ac
Adding inheritance to user defined sampler
...
what if we just do nothing
trying wrapper approach
Fix yolo
4 years ago
fe3b26a637
!12982 dice bug
...
From: @lijiaqi0612
Reviewed-by:
Signed-off-by:
4 years ago
cdb7330818
diceloss bug
4 years ago
57373ed30a
merge getitem the by otehr types to by tuple(success)
4 years ago
e8ee209877
Optimize summary ut and delete duplicate test cases
...
Change the summary test case to stage3 to avoid multiple processes that cause the summary test case to fail
4 years ago
91cf88d692
!12883 Fix bugs of minddata transforms
...
From: @luoyang42
Reviewed-by:
Signed-off-by:
4 years ago
74431ad117
Fix bugs of minddata transforms
4 years ago
7ba21f8d8c
!12900 Add communication parallel mode.
...
From: @liujunzhu
Reviewed-by: @zhoufeng54,@guoqi1024
Signed-off-by: @guoqi1024
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
fe53c71b3b
!12843 Add more callable test and clean the input check for tensor ops
...
From: @alexyuyue
Reviewed-by: @robingrosman,@nsyca
Signed-off-by: @robingrosman
4 years ago
6541b96c40
Add communication parallel mode.
4 years ago
d285692217
!12852 Change maketuple in coreops
...
From: @liangzhibo
Reviewed-by:
Signed-off-by:
4 years ago
680324f225
Change make tuple in core.ops
4 years ago
ec9793861f
fix grad accu
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
7ff2b3b499
!12781 fix bug of amp bn cast
...
From: @jojobugfree
Reviewed-by:
Signed-off-by:
4 years ago
6d7be32614
Add more callable test and clean the input check
...
Signed-off-by: alex-yuyue <yue.yu1@huawei.com>
4 years ago
1c191a65fb
!12765 Fix numpy input to samplers
...
From: @hfarahat
Reviewed-by:
Signed-off-by:
4 years ago
c07371f5c9
!12763 Fix C++ API testcases
...
From: @ezphlow
Reviewed-by:
Signed-off-by:
4 years ago
8e13667ff4
!12674 Add support to multiple tensor in eager mode
...
From: @alexyuyue
Reviewed-by: @nsyca
Signed-off-by:
4 years ago
f9a2379a70
Fixing api leftover
...
Fixed compile error
Fixed more testcases
4 years ago
e58be0de71
!12737 quantum state evolution operator
...
From: @donghufeng
Reviewed-by:
Signed-off-by:
4 years ago
7fcce73c51
!12700 add grad accumulation combined with optimizer parallel
...
From: @yangzhenzhang
Reviewed-by:
Signed-off-by:
4 years ago
8fb01aabd4
!12738 Add heartbeat check in summary and delete test cases that do not exit
...
From: @ouwenchang
Reviewed-by: @yelihua
Signed-off-by:
4 years ago
66aff93940
Fix test team issue: ToNumber in eager mode.
...
And support multiple tensor as input in call method
Signed-off-by: alex-yuyue <yue.yu1@huawei.com>
4 years ago
a6959c2a13
fix bn cast bug
4 years ago
4309e20b1b
Add heartbeat check in summary and delete test cases that do not exit
4 years ago
616d1a1f3d
Fix sampler supporting numpy input
4 years ago
85461bcdb3
!12532 add SyncBatchNorm
...
From: @yuchaojie
Reviewed-by:
Signed-off-by:
4 years ago
066d23e2ad
evolution operator
4 years ago
a70d616841
mini step grad accumulation
4 years ago
f0b7c3dd4d
Add a switch to control grad for scalar
4 years ago
6d195f340c
add SyncBatchNorm
4 years ago
mindspore-ci-bot
buxue
YangLuo
Zirui Wu
alex-yuyue
RobinGrosman
liuxiao93
Eric
jiaqi
yepei6
ougongchang
Zhang Qinghua
liujunzhu
l00591931
Ziyan
caifubi
hesham
xuxs
yangzhenzhang
yujianfeng
yuchaojie