search

xinntao / ESRGAN

ECCV18 Workshops - Enhanced SRGAN. Champion PIRM Challenge on Perceptual Super-Resolution. The training codes are in BasicSR.
https://github.com/xinntao/BasicSR
Apache License 2.0
5.91k stars 1.05k forks source link

Training a new model with different scaling factor #52

Open chhsieh0309 opened 5 years ago

chhsieh0309 commented 5 years ago

Hi there,

If I want to train a new ESRGAN model, a 2.5x upsampling/downsampling factor, for example, how do I get the corresponding pre-trained PSNR model?

xinntao commented 5 years ago

I think you can use the weights of pre-trained X4 ESRGAN to initialize the X2.5 model, except the upsampling layer.

zhangcl commented 4 years ago

@chhsieh0309 have you successfully trained the model with 2.5x scale factor? I was able to train with scale factor of integer values (e.g. 2x, 4x), but not sure how to train a model with non-integer scale factor (e.g. 1.5x, 2.5x).

jorjiang commented 2 years ago

@chhsieh0309 have you successfully trained the model with 2.5x scale factor? I was able to train with scale factor of integer values (e.g. 2x, 4x), but not sure how to train a model with non-integer scale factor (e.g. 1.5x, 2.5x).

why not just use 4x and then downsample to 2.5