ougongchang
9229f1c1ff
profiler support to collect parallel strategy info
If SetNodeOutputType functions forcibly splits into multiple functions, the readability decreases, so it blocks lizard scans
4 years ago
yelihua
cfa8c7a0e8
update OWNERS
4 years ago
yanghaitao1
a30b940ad3
add commiter to OWNERS files
4 years ago
i-robot
9f92ac51b1
!24250 fix some review issues in profiler
Merge pull request !24250 from yanghaitao/yht_fix_review_issue_0927
4 years ago
yanghaitao1
09e2d4bb88
fix some review issues in profiler
4 years ago
yanghaitao1
6a92b2c538
fix reviews issues in profiler
4 years ago
i-robot
a009c8e8b4
!23542 fix pclint in profiler
Merge pull request !23542 from yanghaitao/yht_fix_issue_0915
4 years ago
yanghaitao1
56fd6f57e7
fix pclint in profiler
4 years ago
yanghaitao1
177f3f75bf
remove profiler if compiled with -s on
4 years ago
i-robot
94f7cba97c
!23219 remove profiling parameters in set_context function
Merge pull request !23219 from yanghaitao/yht_remove_set_context_from_profiling_3
4 years ago
zhangyihui
e121fb2e8c
Heterogeneous cpu performance in GPU
4 years ago
yanghaitao1
83302ad23c
remove profiling parameters in set_context function
4 years ago
i-robot
18078a6457
!22917 If users need GPU profiling, they need to use the root account to start the process
Merge pull request !22917 from zangqx/profiling_gpu_permission
4 years ago
臧庆香
2bc81c48e0
profiling GPU permission
4 years ago
zhangyihui
6a36171a0e
transform device_id to rank_id for cpu_profiler
4 years ago
zhangyihui
3d19949eb4
device_id to rank_id
4 years ago
ms_yan
36a8886ca2
Revert "[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset"
This reverts commit b077aa1cab .
Revert "[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset"
This reverts commit 4e6f7dc97d .
delete pass_registry_test.cc
comment hiai_nlu_model_multi.pb related line
4 years ago
djc
b077aa1cab
[feat] [assistant] [I3T96T] add new Dataset operator CMUARCTICDataset
4 years ago
djc
4e6f7dc97d
[feat] [assistant] [I3T96X] add new Dataset operator LibriSpeechDataset
4 years ago
lvchangquan
e8d9803258
add profiling codes based on ascend and gpu in pynative mode
4 years ago
gaojing
310841bd51
profiler cleancode
4 years ago
gzhcv
96dfea1e6f
Add scope-level flops in ascend profiler
4 years ago
gzhcv
4c78a47184
Fit Gpu LoopCount for profiler module v2
4 years ago
limingqi107
406c252834
enable mindRT
4 years ago
gzhcv
001985ca25
Fit Gpu LoopCount for profiler module
4 years ago
yanghaitao1
71990c836a
fix review issue
4 years ago
i-robot
c4581dfd6a
!18033 fix pclint in profiler
Merge pull request !18033 from yanghaitao/yht_fix_warning_r1.2_0609
4 years ago
i-robot
340d529877
!17742 Add profile_memory option to Profiler
Merge pull request !17742 from yanghaitao/yht_add_option_to_memory_2
4 years ago
yanghaitao1
158d3c7a5f
Add profiler owerns
4 years ago
yanghaitao1
b8d83c5c64
fix pclint in profiler
4 years ago
yanghaitao1
09aa4cf4d8
add memory profiling option to profiler
4 years ago
caifubi
61efa2c23b
remove some code
4 years ago
yanghaitao1
127e4d4068
fix profiler pclint&codex
4 years ago
limingqi107
21d168e980
optimize the performance of actor runtimie dynamic shape
4 years ago
gzhcv
9c537a8573
Fix the bug of gpu profiler get device_id incorrectly
5 years ago
gzhcv
96cf55b990
Fix the bug of incorrect op time info
5 years ago
mindspore-ci-bot
ec439fdb97
!12503 refactor the c++ code of gpu profiler
From: @gzhcv
Reviewed-by:
Signed-off-by:
5 years ago
gzhcv
c3a9fe6d06
refactor the c++ code of gpu profiler
Some function from ccsrc/profiler/device/cpu/cpu_data_saver* and ccsrc/profiler/device/gpu/data_saver* could be reused. Some function from ccsrc/profiler/device/cpu/cpu_profiling* and ccsrc/profiler/device/gpu/gpu_profiling* could be reused. This pr is to simplify these code.
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
mindspore-ci-bot
2cfd913e2b
!11830 Add cpu operator profiling and heterogeneous timeline
From: @gzhcv
Reviewed-by:
Signed-off-by:
5 years ago
zhangyihui
bbbffbb6a9
Modify the profiling_options parameter name
5 years ago
gzhcv
2617f6c620
Add cpu profiler
5 years ago
mindspore-ci-bot
d638a92946
!11892 remove unnecessary profiling_output_path parameter for set_context
From: @yanghaitao1
Reviewed-by: @yelihua
Signed-off-by:
5 years ago
yanghaitao1
eefb4295ba
remove profiling output path from context
5 years ago
gzhcv
ccc9ea0317
Fix the bug of step_trace cannot recognize correct fp/bp node in train_eval 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
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
gzhcv
3a40bfc01d
fix the bug of Profiler module do not print meaningful error when training process unexpectedly stopped
5 years ago
gzhcv
de3a653c57
fix the bug of step_trace incorrect content in inference scene
5 years ago