Commit Graph

37 Commits

Author SHA1 Message Date
candanzg 2cc85bdc93 Support weight compile according to shape
Signed-off-by: candanzg <zhangshucheng@huawei.com>
2020-05-13 17:49:07 +08:00
jinyaohui 26fd75895d pylint waring clean 2020-05-13 11:30:27 +08:00
mindspore-ci-bot 66667d727e !1034 Gpu Support Dropout operator
Merge pull request !1034 from SanjayChan/dropout
2020-05-12 09:11:30 +08:00
chenzomi 661f9dfaf8 add dropout primtive 2020-05-11 19:29:18 +08:00
mindspore-ci-bot 8a45ab1125 !906 fix a bug that support dilation greater than 1 in conv2dTranspose ops
Merge pull request !906 from yangyongjie/master
2020-05-09 21:58:27 +08:00
mindspore-ci-bot deae380969 !637 Learning rate and weight decay making group params
Merge pull request !637 from ghzl/learning-rate-make-group-mode
2020-05-07 15:34:10 +08:00
guohongzilong 824bc30a94 learning rate and weight decay support group mode 2020-05-07 09:38:03 +08:00
mindspore-ci-bot 3d3b9d5474 !840 [BUG] fix conv2dtranspose bug
Merge pull request !840 from jinyaohui/conv2dtranspose
2020-05-06 16:43:28 +08:00
jinyaohui 88e763a98f modify conv2dtranspose 2020-05-06 10:53:36 +08:00
zhaojichen 039c75af8e fix avgpool and add check 2020-05-04 22:28:50 -04:00
yangyongjie 3f1c6b7b47 solve the problem when dialtion greater than 1 in Conv2dTranspose ops 2020-05-04 11:55:18 +08:00
zhaozhenlong 66e7a36846 ImageGradients check 4d
fix DiagPart constant folding issue

fix argmin output type check

fix atan2 doc error

fix remove FusedBatchNorm and its grad
2020-04-29 21:47:33 +08:00
wangnan39@huawei.com 7f602016f4 add parameter verification for rmsprop, and modify default value in annotation 2020-04-29 16:12:58 +08:00
zhaozhenlong c88edfb31d psnr check two input same shape and type 2020-04-26 17:16:58 +08:00
mindspore-ci-bot 818acd46d4 !569 fix bug in checkpoint when save scaler
Merge pull request !569 from wangnan39/master
2020-04-22 18:55:02 +08:00
wangnan39@huawei.com f38d18c665 fix bug in checkpoint when save scaler 2020-04-22 16:04:55 +08:00
mindspore-ci-bot 67057d1309 !541 add average pooling 1D
Merge pull request !541 from JichenZhao/avgpooling
2020-04-22 14:13:04 +08:00
mindspore-ci-bot 72f42fc37c !170 Add prim name to error message for operators in nn_ops.py
Merge pull request !170 from fary86/add_prim_name_to_error_message_for_nn_ops
2020-04-21 21:14:44 +08:00
wangnan39@huawei.com b812b18c02 support update parameter for vm 2020-04-21 20:19:46 +08:00
fary86 6dd72f654a Add prim name to error message for nn_ops.py 2020-04-21 19:25:30 +08:00
zhaojichen 94c99998ae add AvgPooling layer 2020-04-21 03:17:24 -04:00
fary86 8cbbbd950e Add cell name to error message 2020-04-17 12:58:11 +08:00
zhaozhenlong aa8fbcc06e add cell psnr 2020-04-17 09:15:51 +08:00
zhaojichen 04c522d0c6 Add Group Normalization 2020-04-16 01:59:32 -04:00
zhaojichen 0b7de6968f Add Group Normalization 2020-04-16 01:56:12 -04:00
zhaojichen ebe6efff71 Add Group Normalization 2020-04-16 01:42:32 -04:00
zhaozhenlong 6a2cf4b6e6 ssim impl code 2020-04-14 21:53:29 +08:00
mindspore-ci-bot 7ffb8bb19f !250 Add nn.pad to support three modes
Merge pull request !250 from casgj/gaojing_new4
2020-04-14 10:56:05 +08:00
gaojing 2db3e64ff2 add operation 2020-04-13 10:38:03 -04:00
root 7d700295f8 add dynamic lr and enhance optim 2020-04-13 13:48:01 +00:00
chenzomi d64f662c76 quantization aware training frontend operators define. 2020-04-13 12:02:49 +08:00
buxue 7541d3b067 Develop op MaxPoolWithArgMax 2020-04-08 20:47:49 +08:00
zhaozhenlong f9d180d413 add api image gradients 2020-04-08 11:50:22 +08:00
mindspore-ci-bot 352c6faf85 !18 enable use float type learning rate in lars optimizer
Merge pull request !18 from gziyan/master
2020-03-31 18:54:55 +08:00
buxue 0da0bdcf40 Fix bug structure output when there is depend whose first input is constant in outputs 2020-03-30 19:49:46 +08:00
Ziyan 4cbcd8e907 enable use float type learning rate in lars optimizer 2020-03-30 19:05:33 +08:00
zhunaipan 930a1fb0a8 initial version
Signed-off-by: leonwanghui <leon.wanghui@huawei.com>
2020-03-27 22:54:54 +08:00