wang_shaocong
|
fef75213ed
|
Modify implementation of constant_of_shape.
Fix bug of implementation of cast and reduce and infershape of slice and topk.
|
5 years ago |
lyvette
|
ef86629298
|
rewrite aware train converter
|
5 years ago |
zhaozhenlong
|
6d4223bd46
|
add op NonMaxSuppression
|
5 years ago |
mindspore-ci-bot
|
4df56b6c1e
|
!7513 [MSLITE] Add KMeans method for quant
Merge pull request !7513 from ghzl/kmeans-quant
|
5 years ago |
guohongzilong
|
16bfc0bd8b
|
add kmeans method for quant
|
5 years ago |
wangzhe
|
4a71ba1245
|
add layer_norm op and layer_norm fusion
|
5 years ago |
xuanyue
|
7f42991624
|
append onnx parser
|
5 years ago |
Haim Moushkatel
|
9392e02a36
|
Adam optimizer + Bug Fix
|
5 years ago |
yoni
|
bcee320dbb
|
tod add networks and ops
|
5 years ago |
xuanyue
|
357b597b4f
|
add onnx parser and adjust the way of printing unsupport op
|
5 years ago |
xuanyue
|
5a83415f07
|
adjust conv_param initial and add tflite custom parser
|
5 years ago |
zhaozhenlong
|
0e07b0d9e8
|
scale add relu and relu6 for fusion and calculating
|
5 years ago |
xuanyue
|
40cb36efbf
|
fix onnx weightquant and add tflite custom op
|
5 years ago |
xutianchun
|
561ebcca7e
|
promote weight quantization precision
|
5 years ago |
zhaozhenlong
|
9b5400bb6e
|
hard tanh fp32
|
5 years ago |
mindspore-ci-bot
|
b523c9ad53
|
!6434 MSLITE fix equal/gather infershape and onnx deconv parser
Merge pull request !6434 from 徐安越/master
|
5 years ago |
fuzhiye
|
a2c8ce4910
|
solve average pooling avgMode problem
|
5 years ago |
xuanyue
|
8e3fe59f24
|
fix bugs
|
5 years ago |
zhaozhenlong
|
8946acc35b
|
add coffe for reduce
|
5 years ago |
xuanyue
|
6341fa92bd
|
add tflite hashtable_lookup,mirror_pad,skipGram parser;adjust const fold adapt to datatype
|
5 years ago |
xuanyue
|
1b4b287a1c
|
errorcode return and add reduce parser for caffe,lsh_projection for tflite,fix some bugs
|
5 years ago |
mindspore-ci-bot
|
9562d2d5cc
|
!6029 l2norm trainling axis support and multithread
Merge pull request !6029 from zhaozhenlong/lite/issue/l2norm_multithread_trailing_axis
|
5 years ago |
chenjianping
|
6e1ad28724
|
support neg,neg_grad,log_grad
|
5 years ago |
mindspore-ci-bot
|
75cc2b6ed6
|
!5843 mslite fix onnx conv ,reshape,dropout and matmul parser, fix infershape_pass bug
Merge pull request !5843 from 徐安越/master
|
5 years ago |
zhaozhenlong
|
9b4cce8208
|
l2norm multi thread and trailing axis
|
5 years ago |
xuanyue
|
a4aa06db53
|
mslite fix onnx conv\reshape\dropout\matmul parser, fix infershape_pass bug
|
5 years ago |
mengyuanli
|
4d418a7eb1
|
add black_op schema
|
5 years ago |
kai00
|
fe2911021f
|
grad passer add
|
5 years ago |
kai00
|
fc1d615f45
|
grad ops add
|
5 years ago |
mindspore-ci-bot
|
9cb1ae8ab9
|
!5776 support caffe exp op
Merge pull request !5776 from 陶云浩/master
|
5 years ago |
tao_yunhao
|
2e68873a99
|
add arm cpu fp32 op: exp
|
5 years ago |
yoni
|
b601603b55
|
train on device
|
5 years ago |
zhaodezan
|
b6493c209f
|
schema add version
|
5 years ago |
xuanyue
|
1140db795c
|
adjust tflite parser reverse_sequence
|
5 years ago |
wangzhe
|
3616f4c5ef
|
add DetectionPostProcess op
|
5 years ago |
mindspore-ci-bot
|
7f250e8663
|
!5656 add recude to end option on reduce
Merge pull request !5656 from zhaozhenlong/lite/op/resize_caffe
|
5 years ago |
xuanyue
|
99d6d7a07a
|
custom parser tflite
|
5 years ago |
zhaozhenlong
|
4b526d2dd8
|
reduce to end
|
5 years ago |
xuanyue
|
3f3fb1f18d
|
fix tflite and caffe single operator question
|
5 years ago |
chenjianping
|
34f21226a8
|
caffeprelu rename to prelu
|
5 years ago |
xuanyue
|
7c223f2cc8
|
onnx net and fix bug
|
5 years ago |
songhonglei413
|
2931920fea
|
op_pooling+relu
|
5 years ago |
xuanyue
|
5a66ab1564
|
fix issue
|
5 years ago |
mindspore-ci-bot
|
b5df3b97f6
|
!4588 add op_constantofshape and testcase
Merge pull request !4588 from songhonglei413/roi
|
5 years ago |
sunsuodong
|
676e44a130
|
fix_topkv2_parser
|
5 years ago |
songhonglei413
|
b9e69b2714
|
add op_constantofshape and testcase
|
5 years ago |
cjh9368
|
8a29d90d3c
|
solve fp16tofp32 bug
|
5 years ago |
mindspore-ci-bot
|
4bd40f3677
|
!4483 support fp16 models
Merge pull request !4483 from cjh9368/fp16
|
5 years ago |
wandongdong
|
1112a7d5b9
|
add to_format opencl op
|
5 years ago |
cjh9368
|
f3b794bc26
|
support fp16 model
|
5 years ago |