mindspore-Ecosystem
/
mindspore
21
0
Fork 27
Code Issues Pull Requests 6 Projects Releases Wiki Activity
6,103 Commits 23 Branches 30 Tags 4 GiB
Commit Graph

6103 Commits

Author SHA1 Message Date
mindspore-ci-bot 488d1904b6 !3802 fix nopnode internal output error 
Merge pull request !3802 from kisnwang/fix-ps-task-error
 2020-08-03 10:41:21 +08:00
mindspore-ci-bot 7773eafd11 !3848 add anf fusion pass 
Merge pull request !3848 from zhengjun10/master
 2020-08-03 10:37:51 +08:00
mindspore-ci-bot ccb80694fd !3845 add image2d for opencl depthwise op and testcase 
Merge pull request !3845 from wandongdong/master
 2020-08-03 10:33:15 +08:00
mindspore-ci-bot ad3d490d1e !3839 GPU: inceptionv3 support in modelzoo 
Merge pull request !3839 from hhc2020/gpu-inceptionv3-modelzoo
 2020-08-03 10:24:00 +08:00
mindspore-ci-bot ecb8738524 !3847 modify populate op parameter 
Merge pull request !3847 from yangruoqi713/lite
 2020-08-03 10:14:09 +08:00
mindspore-ci-bot 49ba473bcc !3803 add gpu klDivLoss op 
Merge pull request !3803 from baihuawei/loss
 2020-08-03 10:07:34 +08:00
mindspore-ci-bot 12102ae3e2 !3830 lite gpu opencl convolutoin kernel speed up 
Merge pull request !3830 from 王东旭/master
 2020-08-03 10:03:11 +08:00
yangruoqi713 c65d63a4af modify populate op parameter 2020-08-03 09:59:43 +08:00
hanhaocheng 83580c4dec GPU inceptionv3 support in modelzoo 2020-08-03 09:47:24 +08:00
mindspore-ci-bot 67005d42c8 !3842 Add new hms ops of crop with type of int8 
Merge pull request !3842 from liuwenhao/master
 2020-08-03 09:41:14 +08:00
mindspore-ci-bot aafa76be26 !3849 modify version number to r0.6 for master 
Merge pull request !3849 from changzherui/mod_ver_num
 2020-08-03 09:37:22 +08:00
mindspore-ci-bot 59530248a3 !3795 remove old MINDSPORE_HCCL_CONFIG_PATH in model zoo 2 
Merge pull request !3795 from panbingao/master
 2020-08-03 09:31:49 +08:00
mindspore-ci-bot 51fcaf6e61 !3714 stridedslice/stridedslicegrad 4D to 7D 
Merge pull request !3714 from panbingao/stridedslice
 2020-08-03 09:19:25 +08:00
mindspore-ci-bot f4ac6a2ccb !3600 Modifying SGD operator bug and adding operator check. 
Merge pull request !3600 from linqingke/new_ops
 2020-08-03 09:18:09 +08:00
mindspore-ci-bot 6e21618fe1 !3798 add shuffle switch for finetune dataset 
Merge pull request !3798 from yoonlee666/master-finetune
 2020-08-03 09:16:14 +08:00
mindspore-ci-bot 2b56562770 !3794 modify inference return type 
Merge pull request !3794 from dinghao/master
 2020-08-03 09:06:06 +08:00
mindspore-ci-bot 465521b9d5 !3698 add spacetodepth and spacetobatch fp32 
Merge pull request !3698 from zhongligeng/master_lite
 2020-08-03 08:37:09 +08:00
mindspore-ci-bot 2883f9366d !3738 fix bug of cast dtype when using mix_presion in pynative mode 
Merge pull request !3738 from jinyaohui/master
 2020-08-02 23:29:51 +08:00
changzherui 106ece221d modify version to r0.6 for master 2020-08-02 17:28:09 +08:00
zhongligeng 4b6b1ddc0e spacetodepth and spacetobatch fp32 
dequantize and quantize
 2020-08-02 14:21:46 +08:00
wandongdong 60743492b3 add image2d for depthwise 2020-08-01 21:27:33 +08:00
zhengjun10 1561462925 add anf pass 2020-08-01 20:50:52 +08:00
mindspore-ci-bot 3d7732a7b3 !3726 add lite op PriorBox 
Merge pull request !3726 from zhaozhenlong/lite/op/prior_box
 2020-08-01 17:48:47 +08:00
liuwenhao4 b3cac13a1c Add new hms ops of crop with type of int8 2020-08-01 17:36:04 +08:00
mindspore-ci-bot 7d3f56c54f !3688 Add deconv parser 
Merge pull request !3688 from ghzl/add-deconv-parser
 2020-08-01 17:35:16 +08:00
mindspore-ci-bot 9ba4e4d88d !3844 little optimization for depth wise kernels 
Merge pull request !3844 from lixian/master
 2020-08-01 17:30:27 +08:00
mindspore-ci-bot 9eb0a7697f !3713 detect model fix bug 
Merge pull request !3713 from ghzl/deconv-adapter
 2020-08-01 17:27:53 +08:00
mindspore-ci-bot d7bc28dcde !3835 add lite kernel relu_int8 and hswish_int8 
Merge pull request !3835 from sunsuodong/int8_relu_hswish
 2020-08-01 17:17:03 +08:00
lixian f894ffe6ad little optimization for depth wise kernels 2020-08-01 17:15:57 +08:00
mindspore-ci-bot bab5aedd49 !3843 fix kernel registry bug 
Merge pull request !3843 from fuzhiye/mindspore
 2020-08-01 17:04:36 +08:00
mindspore-ci-bot cf4533da68 !3810 update the performance of concat ops 
Merge pull request !3810 from pengyongrong/master
 2020-08-01 16:59:15 +08:00
fuzhiye 1eb86b7df2 fix kernel registry bug 2020-08-01 16:54:45 +08:00
zhaozhenlong 90553fef08 add lite op PriorBox 2020-08-01 16:53:13 +08:00
pengyongrong 514ba6e976 Optimize performance for master branch 
modify CMakeLists.txt

remove old MINDSPORE_HCCL_CONFIG_PATH in model zoo

decoupling core and debug

add lite/test

add gpu python

pad fp32 support less than 4d input

repair distribute training for vgg in D
 2020-08-01 16:51:10 +08:00
guohongzilong 668db1dd7d deconv adapter 2020-08-01 16:28:57 +08:00
mindspore-ci-bot d1b3096418 !3837 add_test_for_mindrecord 3 
Merge pull request !3837 from liucunwei/add_test_for_mindrecord3
 2020-08-01 16:09:50 +08:00
sunsuodong 64c2f195b7 int8_relu_hswish 2020-08-01 15:49:12 +08:00
mindspore-ci-bot 5338128283 !3833 rewrite op register func 
Merge pull request !3833 from fuzhiye/mindspore
 2020-08-01 15:16:10 +08:00
fuzhiye 3a70fd23ce rewrite op register func 2020-08-01 15:08:43 +08:00
mindspore-ci-bot 5a3b557354 !3814 enhance: map file label start begin 0 in imagenet_to_mr tool 
Merge pull request !3814 from liucunwei/enhance_map_file_for_imagenet
 2020-08-01 15:01:45 +08:00
wangdongxu 04e4cba6f0 update lite gpu opencl convolution kernel 2020-08-01 14:58:51 +08:00
mindspore-ci-bot 11f786c9bd !3808 remove predict 
Merge pull request !3808 from 张学同/to_merge
 2020-08-01 14:55:08 +08:00
mindspore-ci-bot 53c40d0ec7 !3836 1. multithreading support for fc_int8_op. 2. change asm matmul output layout 
Merge pull request !3836 from zhanyuan/master
 2020-08-01 14:36:40 +08:00
mindspore-ci-bot 3819945547 !3776 fix compilation bugs on arm32 & add BiasAdd/Relu/DW kernels on arm64 
Merge pull request !3776 from lixian/master
 2020-08-01 12:02:03 +08:00
z00505269 87668d6ea2 remove predict 2020-08-01 11:45:19 +08:00
liucunwei 8b99114f6d add mindrecord test case 2020-08-01 11:41:58 +08:00
mindspore-ci-bot 6e759cd487 !3815 compute threshold only once in post training quantization 
Merge pull request !3815 from xutianchun/quant_0731
 2020-08-01 11:41:53 +08:00
lixian e52aa2b12f fix compile bugs on arm32 and add BiasAdd/Relu/Dw arm64 kernels 2020-08-01 11:37:16 +08:00
zhanyuan 9cab9f7a91 1. multithreading support for fc_int8_op. 2. change asm matmul output layout from col8x8 to row8x8 2020-08-01 11:35:13 +08:00
xutianchun fae78e11a7 1. compute threshold only once 
2. fix anf_exporter bug: pool, concat op may not set into metagraph
3. fix weight trans pass will return error when post quantization
4. make anf_exporter reentrant: do not set PrimitiveT * to nullptr
 2020-08-01 10:52:46 +08:00