Dataset about deep-multimodal-subspace-clustering-networks

[mahdiabavisani]

Using the dataset for scientific research requires signing an agreement. Please email the points of contact mentioned in the Appendix of the original paper:

http://openaccess.thecvf.com/content_cvpr_2016_workshops/w4/papers/Hu_A_Polarimetric_Thermal_CVPR_2016_paper.pdf

[HuangQinJian]

Using the dataset for scientific research requires signing an agreement. Please email the points of contact mentioned in the Appendix of the original paper:

http://openaccess.thecvf.com/content_cvpr_2016_workshops/w4/papers/Hu_A_Polarimetric_Thermal_CVPR_2016_paper.pdf

Thank!I aslo want to know the input size of minst?

[HuangQinJian]

Using the dataset for scientific research requires signing an agreement. Please email the points of contact mentioned in the Appendix of the original paper:

http://openaccess.thecvf.com/content_cvpr_2016_workshops/w4/papers/Hu_A_Polarimetric_Thermal_CVPR_2016_paper.pdf

Also,when I scale tihe input pixel into 0-1(pixel/255),the result is so bad,why?

[mahdiabavisani]

The input size for MNIST dataset is also 32x32. (pixel's range between 0 and 255).

If you normalize your data to a range between 0 and 1, you may need to find appropriate regularization parameters through cross-validation (reg_constant1 and reg_constant2 in the code.). The current regularization parameters are borrowed from the DSC paper, where their input is ranged between 0 and 255.

Please check the data processing section in the readme.

[HuangQinJian]

The input size for MNIST dataset is also 32x32. (pixel's range between 0 and 255).

If you normalize your data to a range between 0 and 1, you may need to find appropriate regularization parameters through cross-validation (reg_constant1 and reg_constant2 in the code.). The current regularization parameters are borrowed from the DSC paper, where their input is ranged between 0 and 255.

Please check the data processing section in the readme.

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

I did not find the function in python?

[HuangQinJian]

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

the result decreases as the epoch growing,why?

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