i-robot
981eae461a
!30118 自动优化器并行特性
...
Merge pull request !30118 from zhuyuxiao/I4S85V
2022-02-24 11:09:44 +00:00
i-robot
bbcfbce9e0
!29997 [Auto parallel] [MoE] Support data_parallel + expert_parallel
...
Merge pull request !29997 from Xiaoda/124-moe-changes
2022-02-24 09:23:47 +00:00
zhuyuxiao
d0e0e305d3
good
2022-02-24 16:10:18 +08:00
i-robot
f2130e7434
!30483 [AutoParallel]Pipeline Automatic detection Opt
...
Merge pull request !30483 from lichen/pipeline_opt_detection
2022-02-24 03:40:15 +00:00
yao_yf
e21f878e14
adasum ut fix
2022-02-24 09:31:04 +08:00
Xiaoda Zhang
b714451937
implementing expert_parallel+data_parallel in MoE:
...
1) extending _Linear's input as 4-dimension tensor: [outer_batch, expert_dim, -1, hidden], and _Liner's BatchMatMul becomes BatchMatMul(4_dim_tensor, 3_dim_tensor);
2) configuring the _Linear's BatchMatMul sharding strategy as [[dp, ep, 1, 1], [ep, 1, mp]];
3) introducing a new parameter 'expert_parallel' in TransformerOpParallelConfig, creating a new class MoEParallelConfig to include 'data_parallel', 'model_parallel' and 'expert_parallel';
4) changing parallel config for FeedForward, TransformerEncoderLayer, TransformerDecoderLayer.
2022-02-24 09:24:27 +08:00
wangshengnan12@huawei.com
acbefd80ea
pipeline_opt_detection
2022-02-24 09:23:17 +08:00
i-robot
2a00ffd3b1
!30259 [MD]Update set_autotune_enable API to add save filepath
...
Merge pull request !30259 from cathwong/ckw_at_save_api
2022-02-24 01:00:59 +00:00
Cathy Wong
46e223e569
[MD] Update set_autotune_enable API to add save filepath
2022-02-23 14:50:55 -05:00
i-robot
81260a2319
!30466 takedown test_auto_parallel_adasum.py to ensure stability, again
...
Merge pull request !30466 from yanghaoran/master
2022-02-23 09:15:58 +00:00
i-robot
14393503b7
!30431 allreduce allgather fusion
...
Merge pull request !30431 from jiahongQian/master
2022-02-23 08:52:40 +00:00
yanghaoran
71d6b7d506
takedown test_auto_parallel_adasum.py to ensure stability, again
2022-02-23 16:42:13 +08:00
i-robot
2e8eac8341
!30367 auto_parallel_adasum_support_data_parallel
...
Merge pull request !30367 from yao_yf/auto_parallel_adasum_support_data_parallel
2022-02-23 06:50:26 +00:00
jiahongQian
8a2151d8bb
allgather reducescatter fusion
2022-02-23 10:47:30 +08:00
i-robot
eeb731ae3e
!18738 [assistant][ops]New operator implementation, include LSUNDataset
...
Merge pull request !18738 from Wangsong95/lsun_dataset
2022-02-23 02:28:40 +00:00
i-robot
5bee7156b9
!30369 add_virtualdataset_ut
...
Merge pull request !30369 from lilei/add_virtualdataset_ut
2022-02-23 02:21:34 +00:00
i-robot
0f24b679ec
!29819 Add GlobalNorm Search
...
Merge pull request !29819 from huangxinjing/add_global_norm_search
2022-02-23 01:15:08 +00:00
i-robot
abb89d3f06
!29705 [MD][Offload] Support Multi-Column Datasets and Map Column Names to Index
...
Merge pull request !29705 from markuskunej/offload_multi_col_ds
2022-02-22 22:48:38 +00:00
i-robot
6edc6fccee
!30189 Optimize error message while outmost network input is wrong
...
Merge pull request !30189 from zhangzhaoju/master_outmost
2022-02-22 13:05:23 +00:00
yao_yf
19236b1a70
auto parallel adasum support data parallel and hybrid parallel
2022-02-22 20:14:00 +08:00
huangxinjing
092ba035e3
Add global norm parallel support
2022-02-22 16:08:53 +08:00
liyong
3e1c668247
fix mnist dataset
2022-02-22 15:25:13 +08:00
yanghaoran
bfe139b662
takedonw test_auto_parallel_adasum.py to ensure gate stability
2022-02-22 14:51:03 +08:00
lilei
bc62e24d94
add_virtualdataset_ut
2022-02-22 11:38:10 +08:00
i-robot
ce5e93593d
!30301 modify error log
...
Merge pull request !30301 from lianliguang/change-error-log
2022-02-22 02:29:48 +00:00
i-robot
94c8c6355c
!30294 auto_parallel_adasum_checks_and_ut.
...
Merge pull request !30294 from yao_yf/auto_parallel_adasum_checks_and_ut
2022-02-22 01:10:24 +00:00
i-robot
926fb6f5ae
!25559 [feat] [assistant] [I48O7X, I48O6R] Add Asin, AsinGrad
...
Merge pull request !25559 from 桂宁馨/Asin
2022-02-21 14:00:58 +00:00
zhangzhaoju
e000889265
Fix issue I4KVTZ
...
The description of "outmost network" in error message instead by
method/function name
2022-02-21 19:52:53 +08:00
i-robot
95f5075371
!30190 [AUTO_PARALLEL]Fix CI error
...
Merge pull request !30190 from huangxinjing/fix_ci_error
2022-02-21 11:50:28 +00:00
lianliguang
efabace125
change error log
2022-02-21 17:48:16 +08:00
i-robot
908beb6b2f
!26420 [feat] [assistant] [I48O90, I48O4P] Add AsinhGrad,Asinh
...
Merge pull request !26420 from 桂宁馨/Asinh
2022-02-21 09:33:18 +00:00
i-robot
f6215c614c
!30064 modify ckpt type error
...
Merge pull request !30064 from changzherui/mod_ckpt_type_err
2022-02-21 08:59:00 +00:00
huangxinjing
5e325ac336
[AUTO_PARALLEL]Fix insert nodes error
2022-02-21 16:39:50 +08:00
i-robot
a5242d33e1
!30181 support TensorShape ops
...
Merge pull request !30181 from lianliguang/change-dynshape-to-tensorshape
2022-02-21 07:36:40 +00:00
i-robot
7386612515
!29820 moe_topk routing
...
Merge pull request !29820 from wangshengnan123/moe_topk_routing
2022-02-21 06:21:55 +00:00
yao_yf
4b79d4c425
auto parallel adasum uts and checks
2022-02-21 10:41:09 +08:00
lianliguang
5c8af66e53
support TensorShape ops to instead of DynamicShape ops
2022-02-21 09:33:32 +08:00
markuskunej
87a45b239b
Add offload support for multi-column datasets and map offload operations
...
to specific columns.
2022-02-18 20:09:49 +00:00
zhixinaa
4ce6104ff9
[feat][assistant][I3J6V9] add new data operator LSUN
2022-02-18 15:40:32 +08:00
桂宁馨
caf54a52f0
[feat] [assistant] [I48O90, I48O4P] Add Asinh, AsinhGrad
2022-02-18 15:08:55 +08:00
桂宁馨
911c6eb144
[feat] [assistant] [I48O7X, I48O6R] Add Asin, AsinGrad
2022-02-18 14:11:41 +08:00
wangshengnan123
7322426648
top_k routing
2022-02-18 14:10:22 +08:00
changzherui
05dca07cf6
modify ckpt type error
2022-02-18 00:42:55 +08:00
i-robot
48d4f34576
!30167 Add UT case for 'Convolution+Transformer' structure
...
Merge pull request !30167 from Bert0108/ut_conformer
2022-02-17 07:43:58 +00:00
i-robot
7c32002e0b
!29930 Support List Slice
...
Merge pull request !29930 from lianliguang/support-list-slice
2022-02-17 07:24:40 +00:00
Bert0108
25a9c73a08
add ut case for monitoring the conformer structure
2022-02-17 15:08:01 +08:00
i-robot
465f21a3ea
!30133 Add backend check for RandomChoiceWithMask
...
Merge pull request !30133 from liuluobin/rnd_fix_master
2022-02-17 02:11:54 +00:00
lianliguang
49608835c1
support list slice
2022-02-17 09:47:26 +08:00
i-robot
c4c5a1c6da
!29862 MD Autotune supporting MindRecordOp
...
Merge pull request !29862 from Mohammad Motallebi/autotune_addworker_mindrecord
2022-02-16 23:34:11 +00:00
mohammad
ddd3b07648
MD autotune supports mindrecord
2022-02-16 14:00:31 -05:00