mindspore-ci-bot
defcc51641
!13304 refactor RDR to support single name
From: @luopengting
Reviewed-by: @ouwenchang,@lixiaohui33
Signed-off-by: @lixiaohui33
5 years ago
luopengting
c8ba7694c5
refactor RDR to support single name
1. support single name
2. add hash method for pair
3. move constructor and destructor of MemAddressInfo as public
4. remove graph_id
5. modify interval for somas info
5 years ago
mindspore-ci-bot
eb1c0310a9
!13307 GPU fix shared_ptr in GpuKernel
From: @VectorSL
Reviewed-by: @cristoval,@chujinjin
Signed-off-by: @chujinjin
5 years ago
mindspore-ci-bot
77cda67b3f
!13012 add mul fusion based on allreduce fusion
From: @lvchangquan
Reviewed-by:
Signed-off-by:
5 years ago
lvchangquan
31f9e6a42c
add op_mul fusion based on allreduce fusion in pynative mode
5 years ago
VectorSL
36e11ae17c
fix GPUKernelMod about the using of shard_ptr
5 years ago
limingqi107
a046a5eb43
optimize GPU format transform
5 years ago
mindspore-ci-bot
a21c8e13b5
!13010 Add device id log
From: @zpac
Reviewed-by: @cristoval,@wilfchen
Signed-off-by: @cristoval
5 years ago
tanghuikang
dac64f30ee
Support ms_function + heterogenous
5 years ago
ZPaC
f2edee750a
Add device id log
5 years ago
wenfangpei
d6b3a07b4a
parallel build gpu ops about graph kernel
5 years ago
mindspore-ci-bot
2f312dac66
!12091 Performance optimization for PyNative AllReduce
From: @jojobugfree
Reviewed-by:
Signed-off-by:
5 years ago
mindspore-ci-bot
4365c332e6
!12813 unify AvgPoolGrad's MindIR
From: @yuchaojie
Reviewed-by: @kisnwang
Signed-off-by:
5 years ago
yuchaojie
d2cb3aa1c2
unify AvgPoolGrad
5 years ago
louei5
99203038a5
support recording gpu memory information and graph execute order
5 years ago
caifubi
171b468bb3
PyNative AllReduce Bucket
5 years ago
mindspore-ci-bot
50542793c8
!12077 optimize gpu backend logger
From: @wilfchen
Reviewed-by: @cristoval,@limingqi107
Signed-off-by: @limingqi107
5 years ago
wilfChen
58196f1faf
modify gpu backend logger
5 years ago
zuochuanyong
3fa26683ac
nlp perf(Pynative): change memory sync mode from synchronous to asynchronous in SyncHostToDevice
5 years ago
He Wei
7d9a783993
[auto-monad] Support side-effects by auto-monad
The basic idea is: exploits data dependency to control the execution order
of side-effect operations, and keep the semantics of ANF unchanged.
The ControlDepend primitive is removed and there are two primitives added:
1. UpdateState:
```
a = Assign(para, value)
```
became:
```
a = Assign(para, value, u)
u = UpdateState(u, a)
```
2. Load:
```
x = Add(para, value)
```
became:
```
p = Load(para, u)
x = Add(p, value)
u = UpdateState(u, p)
```
5 years ago
lizhenyu
6649153c49
add input data type check for ps cache mode
5 years ago
limingqi107
366f3e668d
optimize the memory alloc error info
5 years ago
mindspore-ci-bot
4364abc7ee
!11798 Support RunOpsInGraph on CPU&GPU in pynative mode
From: @HulkTang
Reviewed-by:
Signed-off-by:
5 years ago
lizhenyu
f17534af08
ps cache support sparse
5 years ago
tanghuikang
6f2cd92aba
Support RunOpsInGraph on CPU&GPU in pynative mode
5 years ago
mindspore-ci-bot
03f88c6f44
!11271 Fix the bug of step_trace cannot get the step_trace_point name in callback scene
From: @gzhcv
Reviewed-by: @ouwenchang,@lilongfei15
Signed-off-by: @lilongfei15
5 years ago
mindspore-ci-bot
d8323b5d51
!11342 Support device memeory profiling
From: @yanghaitao1
Reviewed-by: @wangyue01,@lilongfei15
Signed-off-by: @lilongfei15
5 years ago
yuchaojie
1932d87a26
update some op's attr name
5 years ago
yanghaitao1
8d147deb07
profiler memory
5 years ago
gzhcv
ce66e8cf4e
Fix the bug of step_trace cannot get the step_trace_point name in callback scene
5 years ago
chujinjin
ade9a82c2b
fix device memory leak
5 years ago
mindspore-ci-bot
59d2affa97
!10021 Fixed double stutter issue at the end of step for debugger
From: @adelshafiei
Reviewed-by: @john_tzanakakis,@wangyue01
Signed-off-by: @wangyue01
5 years ago
caifubi
4aecf539e9
Fix GPU sync stream Segmentation fault
5 years ago
John Tzanakakis
b291861e85
LoadTensors was using a 4-byte size field instead of an 8-byte size field
don't load tensors if the type is none
5 years ago
ZPaC
bf793cd1a4
Return false if Recv node is not found in gpu_stream_assign.cc
5 years ago
lizhenyu
1f99cd7d86
ps cache data process thread support exit when exceptions occur
5 years ago
Adel Shafiei
1d795a4f9d
Fixed the double stutter problem at the end of step for debugger
5 years ago
mindspore-ci-bot
726a4b2e8f
!9665 fix the bug of step_trace incorrect content in inference scene
From: @gzhcv
Reviewed-by:
Signed-off-by:
5 years ago
mindspore-ci-bot
bccf04d36c
!8378 [MD][GPU] minddata device_que push optimizer
From: @xiefangqi
Reviewed-by:
Signed-off-by:
5 years ago
xiefangqi
ae36071fc8
add push opt logic
5 years ago
limingqi107
72a584b7c7
gpu all-reduce memory alloc fixed
5 years ago
gzhcv
de3a653c57
fix the bug of step_trace incorrect content in inference scene
5 years ago
lizhenyu
e3f7ae61db
add ps cache manager
5 years ago
mindspore-ci-bot
2799b6d35f
!9683 [Debugger] Performance and state improvements
From: @harsh1995
Reviewed-by: @john_tzanakakis,@wenkai_dist
Signed-off-by: @wenkai_dist
5 years ago
Harshvardhan Gupta
dd0084c52b
improve perf, keep consistent tensor state, fix recheck, check weights at step end
5 years ago
limingqi107
a5f57ce8a0
add ps cache
5 years ago
mindspore-ci-bot
a84f2d812a
!8989 [PyNative]Optimize performance of Resnet50
From: @jojobugfree
Reviewed-by: @jjfeing,@chujinjin
Signed-off-by: @chujinjin
5 years ago
mindspore-ci-bot
1a5dd4a711
!9390 Pynative support dynamic op run in gpu
From: @joylvliang
Reviewed-by: @chujinjin,@jjfeing
Signed-off-by: @chujinjin
5 years ago
mindspore-ci-bot
95573571f0
!9511 Codedex change for tensor_loader
From: @liangzhibo
Reviewed-by:
Signed-off-by:
5 years ago
lvliang
8984cc9c03
pynative-support-dynamic-op-run-in-gpu
5 years ago