yangyan92 / deep-admm-net Goto Github PK

Hi, I tried to run your code under Mac OX and Ubuntu, but both OS fail to work.
I wonder it is have to be run under Windows?

Dear Teacher Yang:
Thank you for your excellent research work,
I have a question to ask about different stages in './net/network_20'.
How do I generate different number of stages, such as 12 stages :'net-stage12.mat'.
In other words, I want to generate something different from what you provide ‘net-stage5.mat, net-stage14.mat, net-stage15.mat’ under the file you provided .
Looking forward to your reply.
Sincerely

It seems that there is no brain training data and chest testing data. Could you make it publishable?
Thank you so much!

Dear Dr. Yang:
I found this is a very interesting work;
I have tried to train network from scratch to reproduce the results displayed in the paper.
But the training process often terminated unexpectedly.
It seems one should modify main_netTrain.m and config.m to make the training successful.
I have changed nnconfig.TrainNumber = 1; in config.m to nnconfig.TrainNumber = 100;; This seems to work, but the training process can still be finished unexpectedly.
So I was wondering what else I should modify to train the network from scratch?

Best wishes