266e960acb
Not do cse for the nodes set recomputed before recompute pass
4 years ago
48f2d82c0e
!11577 Use ThreadPool in ParallelFor
...
From: @wuxuejian
Reviewed-by: @kisnwang,@guoqi1024,@c_34
Signed-off-by: @c_34
4 years ago
19fc6306d3
fix for dataset input
4 years ago
5280ab355e
!11618 [MS_LITE] control flow encoder
...
From: @YeFeng_24
Reviewed-by:
Signed-off-by:
4 years ago
6500655989
!11655 [AutoParallel] log as INFO instead of EXCEPTION for unexpected output tensor shape cases
...
From: @ch-l
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
4 years ago
49e9b80328
nested loop expand
4 years ago
ecca62d786
!11534 profiler: optimized the code of processing memory usage and added memory breakdowns for each execution id
...
From: @zhangyunshu
Reviewed-by:
Signed-off-by:
4 years ago
52707ba76e
!10916 Add some logs for collecting graph and using summary operators
...
From: @ouwenchang
Reviewed-by:
Signed-off-by:
4 years ago
f552d84a94
modify export file name check
4 years ago
26725e81d1
4
4 years ago
afce3e062e
solve undefined symbol error when akg is loaded in ubuntu arm environment
4 years ago
74c2b957f7
!11632 fix gru net usablity bug
...
From: @qujianwei
Reviewed-by: @linqingke,@liangchenghui
Signed-off-by: @liangchenghui
4 years ago
01b77c448a
!11525 [MS][LITE][CPU]optimize reshape,squeeze,tile,strideslice op
...
From: @fuzhiye
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhang_xue_tong
4 years ago
10a12ac159
!11553 [MSLITE][Develop] optimize sigmoid and tanh
...
From: @sunsuodong
Reviewed-by: @zhang_xue_tong
Signed-off-by:
4 years ago
5d7682cef1
!11441 Add F1 score of deeptext's evaluation
...
From: @zhouyaqiang0
Reviewed-by: @c_34,@linqingke
Signed-off-by: @linqingke
4 years ago
0758a95d4e
Adjust the output of MD simulation
4 years ago
9fb9b3b166
!11629 fix shufflenet & googlenet bugs
...
From: @pandoublefeng
Reviewed-by: @c_34,@linqingke
Signed-off-by: @c_34
4 years ago
c43a2eb3ba
!11028 [MSLITE] support onnx tiny-yolov3 loop model
...
From: @zhengjun10
Reviewed-by:
Signed-off-by:
4 years ago
9ed1ad1626
!11608 [MS][LITE][DEVELOP]solve single ops converter bug
...
From: @cjh9368
Reviewed-by: @zhanghaibo5
Signed-off-by:
4 years ago
e36e11e095
optimize op performance
4 years ago
9d00e30ed0
!11544 modify export mindir for type tupletensor
...
From: @changzherui
Reviewed-by: @zh_qh,@zhoufeng54
Signed-off-by: @zh_qh
4 years ago
5889e9c0d9
Renamed variables to lower camel case.
...
Renamed illegal variables and functions.
reconverted the nae of function in Lite java files.
modified
4 years ago
6f38e1de85
!11550 fix cast muilti thread bug
...
From: @yeyunpeng2020
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by:
4 years ago
c0a1f88ad0
optimize_sigmoid
4 years ago
6adb82c2c6
gru_fix_bug
4 years ago
9fddc2e1e4
add pipe
4 years ago
a37e917810
support onnx loop model tiny-yolov3
4 years ago
a7016ea735
!11637 unsqueeze slice space-to-depth
...
From: @ling_qiao_min
Reviewed-by: @zhang_xue_tong,@hangangqiang
Signed-off-by: @zhang_xue_tong
4 years ago
74e4cc1876
!11570 [MS][LITE][Develop] GPU modify some problems for prelu
...
From: @pengyongrong
Reviewed-by:
Signed-off-by:
4 years ago
27b4e1653a
Raise akg ReduceSum precision
...
Cast the float16-input to float32 before ReduceSum, and cast back to float16 after ReduceSum.
If the op after this ReduceSum is a casting from float16 to float32, then it can be eliminated.
4 years ago
1cd3b2fe03
resolve issue
4 years ago
f840c8ccae
fix cast multi thread
4 years ago
76b862bde0
modify layernorm and softmax 2021/1/25 monday
4 years ago
16939ee902
fix bug: np.column_stack raises RuntimeError in graph mode
4 years ago
c6e4b0c85f
Add more log when collect graph and use summary operators
...
Fix can not collect input data when batch size is 1 and total step
number is 1
Fixed spelling errors
4 years ago
aa42728e84
profiler mem usage: added memory breakdowns for each execution id
4 years ago
659b5d8e10
!11633 [MS][LITE]fix pet classification android demo title
...
From: @sishuikang
Reviewed-by: @zhanghaibo5,@zhang_xue_tong
Signed-off-by: @zhanghaibo5
4 years ago
e4e5c1231c
modify layer_norm_fusion
4 years ago
727c7a4668
mod_conv2d_pool_add_ctcloss
4 years ago
c9442eccb7
!11593 remove sparse softmax grad node
...
From: @jjfeing
Reviewed-by: @zhoufeng54,@kisnwang
Signed-off-by: @kisnwang
4 years ago
86dfebd837
!11510 [MS][LITE] weight quant support huffman code
...
From: @jianghui58
Reviewed-by:
Signed-off-by:
4 years ago
336ba49ef2
unsqueeze slice space-to-depth matmul
4 years ago
09acd1680f
!11594 [MS][LITE]Fix illegal instruction error on devices like p8 and p20
...
From: @probiotics_53
Reviewed-by: @hangangqiang,@zhang_xue_tong
Signed-off-by: @hangangqiang
4 years ago
16944856f9
!11457 PS cache insert linear
...
From: @gaoyong10
Reviewed-by:
Signed-off-by:
4 years ago
dc70469d4a
!11597 Fix code spell in Bert and CRNN
...
From: @c_34
Reviewed-by: @liangchenghui
Signed-off-by:
4 years ago
70d575c76a
weight quant support huffman code
4 years ago
0cf280b959
fix demo title
4 years ago
be6d310946
Default to memory-only cache server
4 years ago
b97e99b9ce
change a log level to info
4 years ago
029ea5d1f3
!11301 Push sampler to GeneratorOp c++
...
From: @hfarahat
Reviewed-by:
Signed-off-by:
4 years ago
yujianfeng
mindspore-ci-bot
qujianwei
mengyuanli
changzherui
yefeng
wYann
zhangxinfeng3
fuzhiye
liujiahan
sunsuodong
fangzehua
zhengjun10
dayschan
cjh9368
yeyunpeng
Pengyongrong
wangrao
ougongchang
zhangyunshu
wangzhe
wanyiming
ling
jianghui58
hukang hwx963878
Lixia Chen
sheng