8e60cfe800
[MD] C++ api save
4 years ago
b12c8564b5
add condition stop for kmeans
4 years ago
5532de75ef
!7631 fix post training quant with multi-output op
...
Merge pull request !7631 from xutianchun/mulit
4 years ago
b8a6c88323
api
4 years ago
86d0c61bf2
ddp int8
4 years ago
5f94c782df
!7649 add value check for QuantizationAwareTraining's param optimize_option
...
Merge pull request !7649 from yuchaojie/quant3
4 years ago
d3401c1810
!7633 fix resnet50_quant run_train bug
...
Merge pull request !7633 from yuchaojie/quant2
4 years ago
6d4223bd46
add op NonMaxSuppression
4 years ago
b1dbfa643b
[MSLITE][Develop] optimize arm cpu int8 op conv dw 3x3: add assembly arm32
4 years ago
8defa06a5d
!7583 [ME] change some format code.
...
Merge pull request !7583 from chenzhongming/zomi_master
4 years ago
19b10d474f
!7530 [MD] Add divide image proccessing
...
Merge pull request !7530 from jiangzhiwen/lite/image_divide
4 years ago
6fcf10e0b5
!7598 enbale ps mode init in server
...
Merge pull request !7598 from zyli2020/init_in_server
4 years ago
80725441b6
!7518 Added tcp server based on libevent
...
Merge pull request !7518 from anancds/tcp-server
4 years ago
35624350b4
!7632 [MS][LITE][CPU]change scheme for int8 conv(non-sdot)
...
Merge pull request !7632 from fuzhiye/tmp
4 years ago
7525335b61
!7627 【MSLITE】optimize mindspore lite gpu kernel init time
...
Merge pull request !7627 from 王东旭/opencl_conv_remove_codegen
4 years ago
dc1ecdab28
!7623 [Data]Modify the behavior of the lite dataset, the default setting is to compile the dataset and mindspore lite trainging together
...
Merge pull request !7623 from xulei/lite_test0905
4 years ago
f4446c30c8
raise reduce precision bug fix
4 years ago
06a9b4aa37
Commit1_TensorDot
...
linting errors fix
comments cleanup
typo fix in doc
added more tests, fixed some formatting
changed expected dx values for linter
pylint const data fix
4 years ago
6bcde16c81
!7658 Fix cache android build
...
Merge pull request !7658 from EricZ/add_ir_folder
4 years ago
ee684cc750
fix squeeze dimension issue and gradient issue in sampling
4 years ago
40b45d7447
Fix cache IR android build
4 years ago
f6eaeca62c
!7609 Add c++ API for CacheOp
...
Merge pull request !7609 from h.farahat/cache_c++_api
4 years ago
c94162ac94
Add c++ API for dataset Caching
4 years ago
0d5e119fa4
!7483 Add GPU-UniformSampler and nn.UniformCandidateSampler
...
Merge pull request !7483 from 34bunny/GPU-ucs
4 years ago
2c468a358a
!7537 new gpu op for cbg: repeat_elements
...
Merge pull request !7537 from Peilin/cbg-new-gpu-ops-repeat_elements
4 years ago
aa6b6e594a
Add a batch of new operators
4 years ago
0c043d18ef
modify_normal
4 years ago
8f392204be
minddata build with mindspore-lite train
4 years ago
f8fa043f37
add pynative st test_case
4 years ago
bc3e37c8d5
fix post-training quantization mulit-output
4 years ago
f26a3c685b
!7622 [MSLITE] Decoupling primitive problems
...
Merge pull request !7622 from yeyunpeng2020/primitive
4 years ago
4918e404de
decoupling populate_parameter problem
4 years ago
516c53f1f0
modify int as int32_t
4 years ago
17082f0487
add value check for QuantizationAwareTraining's param optimize_option
4 years ago
4df56b6c1e
!7513 [MSLITE] Add KMeans method for quant
...
Merge pull request !7513 from ghzl/kmeans-quant
4 years ago
9f3b4aae02
!7192 [MSLITE]batchmatmul fusion for tflite transformer
...
Merge pull request !7192 from zhengjun10/transformer
4 years ago
b17cc37918
!7635 add test cases for compatibility test
...
Merge pull request !7635 from zhanghaibo/master
4 years ago
f42353d0d3
!7640 [MSLITE] Optimize depthwise convolution 3x3 for arm64 platform
...
Merge pull request !7640 from zhanyuan/tmp
4 years ago
b34b0cda46
!7641 Amend deeplabv3 readme in master
...
Merge pull request !7641 from jiangzhenguang/amend_deeplabv3_readme_in_master
4 years ago
2edd39c20c
!7628 Update ModelZoo README
...
Merge pull request !7628 from liyanliu96/liyan
4 years ago
b802ce7f8f
!7626 fix for se-resnet50 accurancy
...
Merge pull request !7626 from qujianwei/master
4 years ago
f11336a431
transformer batchmatmul fusion
4 years ago
bc1d33c62d
!7636 fix_export_bug
...
Merge pull request !7636 from baiyangfan/mode_export_bug
4 years ago
d54ea04df3
!7630 dynamic shape bug fix
...
Merge pull request !7630 from liubuyu/master
4 years ago
b43cdc5df9
1.change int8 conv which not support sdot scheme (using matmul)
...
2.free reduandent memory in fp16 kernels
4 years ago
5b6d292107
Amend deeplabv3 readme in master.
4 years ago
ae2295d840
add test cases for compatibility test
4 years ago
c498e175b2
[ME] format code
4 years ago
4b810f5ee3
Optimize depthwise convolution 3x3 for arm64 platform
4 years ago
4bc41ac7c7
make ZerosLike parallel care
4 years ago
luoyang
guohongzilong
mindspore-ci-bot
Jiaqi
wangzhe
zhaozhenlong
yangruoqi713
liubuyu
danishnxt
Xun Deng
Eric
hesham
sheng
lilei
xulei2020
lvchangquan
xutianchun
yeyunpeng
anzhengqi
yuchaojie
zhengjun10
fuzhiye
jzg
zhanghaibo5
chenzomi
zhanyuan
Yi Huaijie