search

sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)
MIT License
2.42k stars 668 forks source link

I want to train the custom data set. #265

Closed DaehanKim-Korea closed 4 years ago

DaehanKim-Korea commented 4 years ago

https://github.com/thstkdgus35/EDSR-PyTorch/issues/51#issuecomment-409424414

Follow the above method, create custom.py, and create dataset/CUSTOM/ folders.

However, in the process of importing data, the following errors were encountered:

ValueError: some of the strides of a given numpy array are negative. This is currently not supported, but will be added in future releases.

So, after checking the print, we found that there was a part that was zeroed, such as HR size = (192, 0, 3) during dataset loading. Do you know how to solve it?

DaehanKim-Korea commented 4 years ago

Additionally, the HR (whole) image was converted through opencv. Here's how:

As a result, path/CUSTOM_train_HR/ HR_images path/CUSTOM_train_LR_bicnic/X4/ downsampled (bicubic) HR s(LR s).

Akshaypatil7 commented 4 years ago

Hi, I am too trying on custom data. Did you manage to successfully run training?

DaehanKim-Korea commented 4 years ago

Hi, I am too trying on custom data. Did you manage to successfully run training?

Yes. I am training successfully now.

Akshaypatil7 commented 4 years ago

@koreadhkim how did you solved that error

LCA-0907 commented 3 years ago

@koreadhkim Also ran into this problem, how did you fix it?

shenhaiyoulan commented 3 years ago

@koreadhkim same problem could you give me some suggestions?