Martin Krasser

I know what Maven offline builds are good for, but my question was: can't we require (from users) to do the _first_ build with an existing internet connection? What speaks...

> Online preparation before traveling is not a problem, of course. Manual copying of the Maven cache between laptops would work as well. But these are IMHO exactly the steps...

Seems you're using an RGBA image (4 channels). You need to convert it to an RGB image first (3 channels).

DIV2K bicubic downscaled images have been created with the MATLAB `imresize` function. It is important that you create downscaled versions of Set5 images in the very same way. A Python...

The provided EDSR model in this repository is their single-scale **baseline** model with 16 residual blocks. The reported PSNR of 32.46 is for their EDSR model with 32 residual blocks....

@xin71 you're welcome, glad it was helpful. Regarding images with an odd number of pixels in a dimension, I'd rather crop the largest possible even-by-even image instead of downsampling as...

@leolya wow, thanks for this useful feedback, much appreciated!

It's a typo, it should be `tf.image.resize`. Thanks for clarifying!

Hi @PIjarihd, the caches are created by the data pipeline. See [data.py](https://github.com/krasserm/super-resolution/blob/master/data.py) for details.

`if downgrade in _downgrades_b and scale != 4` seems to be correct. Why do you think it should be `if downgrade in _downgrades_a and scale != 4`.