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
mindspore-ci-bot
a7bfc16aca
!12660 fix gpu bceloss bug
From: @huaweib
Reviewed-by: @zhoufeng54,@chujinjin
Signed-off-by: @chujinjin
5 years ago
mindspore-ci-bot
d8433e9f5c
!12605 Fix GPU ReduceMax/Sum float64 type failure
From: @TFbunny
Reviewed-by: @tom__chen
Signed-off-by:
5 years ago
mindspore-ci-bot
5a9390abf0
!12596 Fix a bug that squaresumall has precision problem in fp16
From: @yuan_shen_zhou
Reviewed-by:
Signed-off-by:
5 years ago
baihuawei
4a3fcc086b
fix bceloss bug
5 years ago
mindspore-ci-bot
be2019c5be
!12628 Add bond, angle and dihedral modules of Sponge
From: @zhangxinfeng3
Reviewed-by:
Signed-off-by:
5 years ago
zhouyuanshen
52b49f0f86
fix a bug for squaresumall in fp16
5 years ago
TFBunny
ec644a5291
add fix to float64 support in reduce ops
5 years ago
tom__chen
03b1aeecdb
FusedBatchNormEx dynamic shape support
5 years ago
zhangxinfeng3
1bd849462c
Add bond, angle and dihedral modules of Sponge
5 years ago
mindspore-ci-bot
50bf5033f2
!12400 Add float64 support to ZerosLike, remove duplicate ZerosLike gpu kernel
From: @peilin-wang
Reviewed-by:
Signed-off-by:
5 years ago
mindspore-ci-bot
57f6c17933
!12438 Add backward op for IndexAdd GPU
From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by: @robingrosman
5 years ago
Peilin Wang
ba9ae8097f
removed zeroslike from math/unary_op_gpu_kernel and added float64 support
fix ci
fix ci
5 years ago
TFBunny
e1462c8684
add indexaddgrad ops support and testcases
5 years ago
mindspore-ci-bot
cada22e7bf
!12354 Add float64 support to ReduceMax and ReduceSum GPU
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
52eae8d4ad
!12329 GPU index_add forward op
From: @tom__chen
Reviewed-by: @liangchenghui,@robingrosman
Signed-off-by: @liangchenghui
5 years ago
tom__chen
b94bba1def
index_add gpu op
5 years ago
mindspore-ci-bot
7d97d1300e
!12370 Add type support to CumSum gpu op
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
191b3f0c8c
!12368 Add type support to Split gpu op
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
f021af8096
!12367 Add float64 support to MatMul ops in GPU
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
4b27c3206d
!12362 Add float64 support to Abs, Neg, Sqrt, Sin, Cos
From: @peilin-wang
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @robingrosman
5 years ago
chenfei
e48cb95cb7
change float status output type to float32
5 years ago
mindspore-ci-bot
048eba4460
!12359 add float64 support to assign GPU
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
5 years ago
Peilin Wang
5f7fc5a0b1
initial commit
fix ci
5 years ago
Peilin Wang
2252e440a5
initial commit
fix ci
5 years ago
Peilin Wang
073fc6fa93
initial commit
fix ci
5 years ago
TFBunny
4c18e0894e
add float64 support to matmul ops
5 years ago
Peilin Wang
ce26d2e987
intitial commit
fix ci
fix ci
5 years ago
TFBunny
e1ed3b9c15
add float64 support to assign GPU
5 years ago
TFBunny
c6baca71e5
add float64 support to ReduceMax and ReduceSum
5 years ago
mindspore-ci-bot
8d936a6589
!12352 Add float64 support to Gather GPU
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
0ee8ef736d
!12351 Add float64 support to GatherNd GPU
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
5 years ago
mindspore-ci-bot
8fabb26412
!12350 add float64 support to concat GPU
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
5 years ago
TFBunny
6f9cfb809e
add float64 support to GatherNd
5 years ago
TFBunny
6fad165e43
add float64 support to gather gpu
5 years ago
TFBunny
ebb7f2e771
add float64 support to concat GPU
5 years ago
mindspore-ci-bot
d249ededf6
!12347 add float64 support to slice ops in GPU
From: @TFbunny
Reviewed-by: @robingrosman,@tom__chen
Signed-off-by: @tom__chen
5 years ago
mindspore-ci-bot
ed5f9cb1f2
!11802 Optimize GPU kernels for fasterrcnn
From: @robingrosman
Reviewed-by:
Signed-off-by:
5 years ago
TFBunny
799e51cff0
Add float64 support to slice ops
5 years ago
RobinGrosman
5d5225f2ee
optimize fasterrcnn on gpu. fix randomchoicewithmask, topk kernels
5 years ago
mindspore-ci-bot
adf934c567
!12152 Rewrite sequence_mask as a composite op
From: @TFbunny
Reviewed-by: @robingrosman
Signed-off-by:
5 years ago
mindspore-ci-bot
6ed6d57076
!11560 Add dynamic shape support and testcases to GPU Dropout
From: @TFbunny
Reviewed-by:
Signed-off-by:
5 years ago
mindspore-ci-bot
a063d7633d
!12241 [auto-monad] Support side-effects by auto-monad
From: @hwhewei
Reviewed-by: @zhunaipan,@zh_qh
Signed-off-by: @zh_qh
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
CaoJian
811de4f5d4
GPU AbsGrad calculat error while input is 0.0
5 years ago
mindspore-ci-bot
4d9551e129
!12046 modify Gelu、FastGelu to GeLU and FastGeLU
From: @jinyaohui
Reviewed-by:
Signed-off-by:
5 years ago
TFBunny
ad3d737d29
Rewrite sequence_mask as a composite op
5 years ago
mindspore-ci-bot
5a2f472790
!12153 Add type support to StridedSlice GPU
From: @TFbunny
Reviewed-by: @tom__chen,@robingrosman
Signed-off-by: @robingrosman
5 years ago
jinyaohui
30a27b2adb
modify Gelu、FastGelu to GeLU and FastGeLU
5 years ago