About the scirpts to run wavelet cnn and scatttering newtorks

[momo1986]

Very interested in reproducing such results.

If there exist scripts, I am very grateful.

Best Regards! Momo

[mxbastidasr]

Hello Momo,

First of all, thank you for reading my research and for your interest.

There is a github repository with the code.

https://github.com/mxbastidasr/DAWN_WACV2020

Let me know if you have any inconveniences.

Regards,

Ximena

El mié., 6 de ene. de 2021, 1:57 a. m., momo1986 notifications@github.com escribió:

Very interested in reproducing such results.

If there exist scripts, I am very grateful.

Best Regards! Momo

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mxbastidasr/DAWN_WACV2020/issues/2, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFRF4UQB47F2ULVIF754VS3SYQCUFANCNFSM4VXB6RGA .

-- Aviso legal: El contenido de este mensaje y los archivos adjuntos son confidenciales y de uso exclusivo de la Universidad Nacional de Colombia. Se encuentran dirigidos sólo para el uso del destinatario al cual van enviados. La reproducción, lectura y/o copia se encuentran prohibidas a cualquier persona diferente a este y puede ser ilegal. Si usted lo ha recibido por error, infórmenos y elimínelo de su correo. Los Datos Personales serán tratados conforme a la Ley 1581 de 2012 y a nuestra Política de Datos Personales que podrá consultar en la página web  www.unal.edu.co http://www.unal.edu.co/. Las opiniones, informaciones, conclusiones y cualquier otro tipo de dato contenido en este correo electrónico, no relacionados con la actividad de la Universidad Nacional de Colombia, se entenderá como personales y de ninguna manera son avaladas por la Universidad.

[momo1986]

@mxbastidasr . Hello, Ximena.

I have reproduced the program of DAWN.

I am sorry for unclear statement.

I am also interested in reproducing the Wavelet CNN and Scatter networks, thus, I am very sincerely to query you about the scripts to run similar baseline model in the paper.

Thanks & Regards! Momo

[mxbastidasr]

Hello Momo,

In the folder "models" there are the models used to reproduce the work, dawn, densenet, resnet, scattering....

You can choose them as a subparser of the options So for instance, if you want to check the resnet performance: Python train.py --epochs 90 -b 16 --lr 0.01 --database cifar-10 resnet

Regards,

Ximena

El jue., 7 de ene. de 2021, 9:24 a. m., momo1986 notifications@github.com escribió:

@mxbastidasr https://github.com/mxbastidasr . Hello, Ximena.

I have reproduced the program of DAWN.

I am sorry for unclear statement.

I am also interested in reproducing the Wavelet CNN and Scatter networks, thus, I am very sincerely to query you about the scripts to run similar baseline model in the paper.

Thanks & Regards! Momo

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/mxbastidasr/DAWN_WACV2020/issues/2#issuecomment-756147297, or unsubscribe https://github.com/notifications/unsubscribe-auth/AFRF4UTBSUYV2RG7H63CHVTSYW7YHANCNFSM4VXB6RGA .

-- Aviso legal: El contenido de este mensaje y los archivos adjuntos son confidenciales y de uso exclusivo de la Universidad Nacional de Colombia. Se encuentran dirigidos sólo para el uso del destinatario al cual van enviados. La reproducción, lectura y/o copia se encuentran prohibidas a cualquier persona diferente a este y puede ser ilegal. Si usted lo ha recibido por error, infórmenos y elimínelo de su correo. Los Datos Personales serán tratados conforme a la Ley 1581 de 2012 y a nuestra Política de Datos Personales que podrá consultar en la página web  www.unal.edu.co http://www.unal.edu.co/. Las opiniones, informaciones, conclusiones y cualquier otro tipo de dato contenido en este correo electrónico, no relacionados con la actividad de la Universidad Nacional de Colombia, se entenderá como personales y de ninguna manera son avaladas por la Universidad.

[momo1986]

run wcnn has error: python -u train.py --epochs 300 -b 64 --lr 0.03 --name dwnn_cifar-10_32 --lrdecay 150 225 --database cifar-10 --tempdir ~/tmp_cifar_wcnn wcnn --wavelet haar

Could you help double check it if you are convenient?

Thanks & Regards! Momo

[momo1986]

It is cifar-10 dataset and has both error for "scatter" and "wcnn" with such logs:

Traceback (most recent call last): File "train.py", line 942, in main() File "train.py", line 525, in main train_loader, model, criterion, optimizer, epoch, args.model == args_model,scattering=scattering) File "train.py", line 725, in train prec5 = accuracy(output.data, target, topk=(p_m,))[0]#5 File "train.py", line 936, in accuracy correct_k = correct[:k].view(-1).float().sum(0) RuntimeError: view size is not compatible with input tensor's size and stride (at least one dimension spans across two contiguous subspaces). Use .reshape(...) instead.

I try to debug it, it would be my honor if you could give me some directions.

Thanks & Regards! Momo

[momo1986]

Hello, dear Ximena,

I have used contiguous() to make the tensors be contiguous variables to avoid the problems for discrete scenarios.

Thanks for your help.

Maybe this issue can be closed.

With best regards! Momo

[xchhuang]

Hi,

Thanks for releasing various models for testing.

But I had an issue while running the scattering network. It seems that a file utils_scat is missing mentioned in models/scattering.py, would you upload it? Thanks in advance.

[beltegeuse]

Hi,

I have just added utils_scat inside the models\. It was one of the files provided by the authors. This code has been not (re?) tested.

Best.