panyifeng
d6635bbbe2
Add IndexedSlices
5 years ago
Ziyan
0925e35252
enable optimizer parallel with broadcast
5 years ago
gong chen
a6dfa281ea
Init GraphKernel.
- It provides a unified style to express graph and kernel for user.
- It provides a unified IR to represent graph and kernel for developer.
- It breaks the boundary between graph and kernel.
- It provides more opportunities to do compile optimization.
5 years ago
hangangqiang
dfbb232468
mindspore server inference
5 years ago
mindspore-ci-bot
21c67eddd5
!1897 save print data to file
Merge pull request !1897 from jinyaohui/print
5 years ago
jinyaohui
ac23a179bf
save print data to file
5 years ago
Wei Luning
1d77bf86a9
Exports MindSpore quant predict model to deploy with GEIR
5 years ago
chenjianping
6034f9c1e2
support host reduce scatter and mpi config
5 years ago
mindspore-ci-bot
c3d78e2aa2
!2012 gpu support the max device memory config
Merge pull request !2012 from limingqi107/master
5 years ago
limingqi107
5054b53f47
gpu support the max device memory config
5 years ago
Yi Huaijie
e5c351690b
support load full dataset on each device
5 years ago
panyifeng
6a57eeb922
fix check_bprop attr error
5 years ago
z00478463
8c6e3cc83b
for second order's subgraph's fix
5 years ago
z00478463
df5ca0051b
for second order subgraph switch
6 years ago
mindspore-ci-bot
e17c7d2cbc
!1207 add enable_auto_mixed_precision
Merge pull request !1207 from jinyaohui/add_context
6 years ago
kswang
362bbacf19
add group for allreduce fusion
6 years ago
jinyaohui
c0c694d99a
add enable_auto_mixed_precision
6 years ago
jinyaohui
2907cf4488
remove some context param
6 years ago
jinyaohui
391a060f21
remove two context param
6 years ago
jinyaohui
5a9de3716d
add profiling mode and profiling options to context
6 years ago
limingqi107
05e8d95e7f
optimize the gpu context switch
6 years ago
Xiaoda Zhang
def8573275
implementing-searching-strategy-for-inference
6 years ago
mindspore-ci-bot
21d936e656
!728 auto parallel strategy checkpoint full
Merge pull request !728 from yao_yf/strategy_checkpoint_extend
6 years ago
yao_yf
6cde5f6d91
auto parallel strategy checkpoint
6 years ago
zjun
c538b83712
remove enbale hccl
6 years ago
mindspore-ci-bot
e42f7601b6
!315 modify init_ge api name and add init_ge
Merge pull request !315 from jinyaohui/init_dataset
6 years ago
Xiaoda Zhang
e227415673
support-the-multiple-subgraphs-in-the-ANF
6 years ago
jinyaohui
c5156f1e46
add init_device to init_dataset
6 years ago
mindspore-ci-bot
00859ae119
!586 enable/disable allreduce_fusion
Merge pull request !586 from lirongzhen1/allreduce_fusion
6 years ago
Xiaoda Zhang
dfde76af88
delete the 'simplify_cal' attribute in 'set_algo_parameters' and 'get_algo_parameters' interface
6 years ago
lirongzhen
4ff418084c
enable/disable allreduce_fusion
6 years ago
zhoufeng
c2b3360d69
update clang format rule
6 years ago
chenjianping
b6493af6b7
fix gpu init fail
6 years ago
mindspore-ci-bot
b2b3e24a8e
!329 [MS]support building on windows 10
Merge pull request !329 from chenjianping/building-on-windows-2
6 years ago
chenjianping
1286767d0e
support building on windows
6 years ago
Xiaoda Zhang
ffb2cb03a4
Change 'NOT_FULLY_USE_DEVICES' to 'FULLY_USE_DEVICES' and make ALL-1 user-specified-strategy valid in auto-parallel
6 years ago
Wei Luning
73ba399364
remove ge depend in cpu
6 years ago
zhunaipan
930a1fb0a8
initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
6 years ago