AojunZhou Incremental-Network-Quantization issues

AojunZhou / Incremental-Network-Quantization

Caffe Implementation for Incremental network quantization

Other

191 stars 74 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

How to save the model with low-precision?

#41 fantexibaba opened 5 years ago
1
How to calculate the compression ratio

#40 Biyu-GitHub opened 5 years ago
0
Pre-trained networks not available for download

#39 paulsc96 opened 5 years ago
0
weights are not all 2^-n value

#38 xiao777dong opened 5 years ago
1
Weights not being quantized?

#37 yy665 opened 5 years ago
0
Why does the quantized file size remain unchanged？

#36 fantexibaba opened 5 years ago
8
I wonder why you put the INQ operation in caffe key source code like blob.cpp not to create new layers

#35 nejyeah closed 6 years ago
1
how to save int weights to caffemodel ?

#34 wuzhiyang2016 opened 6 years ago
0
For CuDNN higher version support

#33 mdamircoder opened 6 years ago
0
Does n-bits quantization with a largerr n make sense in your methods?

#32 XiangyuWu opened 6 years ago
9
runtest failed (68 FAILED TESTS)

#31 mdamircoder closed 6 years ago
1
make all failed : libcaffe.so.1.0.0-rc3

#30 mdamircoder closed 6 years ago
1
you only quantizatize the weights ?

#29 victorygogogo opened 6 years ago
2
Cannot replicate the results of the INQ paper + Suggestion to optimize the speed of quantization

#28 vassiliad opened 6 years ago
5
Has anyone successfully gain accuracy after quantification on vgg16?

#27 blueardour opened 6 years ago
4
How to set accumulated portions in the code?

#26 WeixiangXu closed 6 years ago
0
Make Error. Does This Caffe Support CUDA8.0-CUDNN7.5?

#25 AilvenLiu closed 6 years ago
0
The caffemodel in your paper will be share?

#24 Ai-is-light opened 6 years ago
0
where is the ‘check.py‘’ file？

#23 simonzeus opened 6 years ago
2
How to change 5 bits to 4,3, 2 bits?

#22 csyhhu opened 6 years ago
1
How to implement gradient descent with mask?

#21 xmfbit closed 6 years ago
2
Baidu caffemodel library missing

#20 briansune opened 6 years ago
0
Effectiveness of Detection Tasks and Tiny Models?

#19 power0341 opened 6 years ago
7
The weight is too large.

#18 lamperouge11 opened 6 years ago
1
how to handle the weight‘s sign in bit-shift arithmetic？

#17 KangolHsu opened 6 years ago
1
train initialization soooo slowly, is it normal?

#16 VivienFu opened 6 years ago
0
can we transform weight to {0,1,2,4,8,...}?

#15 zlheos opened 6 years ago
0
when can we see low-precision activations ?

#14 zlheos opened 6 years ago
0
about the value of weights

#13 KangolHsu opened 6 years ago
5
why use 1 bit to store 0

#12 DAVIDNEWGATE opened 6 years ago
2
What is your original caffe git commit sha-1? Thank you very much

#11 snowygoose opened 6 years ago
0
what is mask_ in blob.cpp?

#10 TwistedfateKing opened 6 years ago
5
ignoring input

#9 muhammadzulfanazhari closed 6 years ago
0
Can this method accelerate inference speed?

#8 mynameischaos opened 6 years ago
3
Bias term in the Convolution param

#7 TwistedfateKing opened 6 years ago
1
bit-width adjustment Question.

#6 TwistedfateKing opened 6 years ago
7
Have you release the script ‘check.py’？

#5 KangolHsu opened 6 years ago
0
Where is the code about setting the partial propagation of weights in a specific layer？

#4 KangolHsu closed 6 years ago
1
How can I reduce the model size?

#3 mychina75 opened 6 years ago
2
Combination of pruning and INQ

#2 DAVIDNEWGATE opened 7 years ago
18
How can I run your code?

#1 TomatoScrambledEggs opened 7 years ago
2