search

mahdiabavisani / Deep-multimodal-subspace-clustering-networks

Tensorflow implementation of "Deep Multimodal Subspace Clustering Networks"
https://arxiv.org/abs/1804.06498
70 stars 23 forks source link

need help #4

Closed HuangQinJian closed 5 years ago

HuangQinJian commented 5 years ago

I train on the digit dataset,the result is bad: image

the result decreases as the epoch growing,why?

I have checked the dataset and the network,and the vision of restruction picture is good image

Originally posted by @HuangQinJian in https://github.com/mahdiabavisani/Deep-multimodal-subspace-clustering-networks/issues/3#issuecomment-500281116

HuangQinJian commented 5 years ago

Thank！Also, Could you provide the code for plot the picture :

image

I did not find the function in python?

Originally posted by @HuangQinJian in https://github.com/mahdiabavisani/Deep-multimodal-subspace-clustering-networks/issues/3#issuecomment-500266300

HuangQinJian commented 5 years ago

I train on the digit dataset,the result is bad: image

the result decreases as the epoch growing,why?

I have checked the dataset and the network,and the vision of restruction picture is good image

Originally posted by @HuangQinJian in #3 (comment)

Could share the digit dataset with me ,so that I can find the problem that I met

mahdiabavisani commented 5 years ago

The coefficient matrix is saved in the variable "C" in a .mat file in the models_DSC folder (see the last line of the code). As to the digits, have you changed the kernel size, and hidden units to match the network reported in the paper? Is the number of samples around 2k samples? If not the learning rate or the max_step of iterations should be changed. From the results you posted, it seems that your learning rate is too high. Try a smaller learning rate (maybe 10e-5? ) or smaller max number of iterations (perhaps max_step =100 and display_step = 10) ... .

HuangQinJian commented 5 years ago

The coefficient matrix is saved in the variable "C" in a .mat file in the models_DSC folder (see the last line of the code). As to the digits, have you changed the kernel size, and hidden units to match the network reported in the paper? Is the number of samples around 2k samples? If not the learning rate or the max_step of iterations should be changed. From the results you posted, it seems that your learning rate is too high. Try a smaller learning rate (maybe 10e-5? ) or smaller max number of iterations (perhaps max_step =100 and display_step = 10) ... .

But the figure does not only need coefficient matrix,maybe needs predict label,right?

Also, I redo the setting as you said above and all other are the same as reported in the paper,it still went bad.

Could share the digit dataset with me ,so that I can find the problem that I met.