yulunzhang
Could you kindly provide the learning curves?
Thank you for your great work! As I don't have enough GPUs to train the model for 1000 epochs, much fewer epochs are used for my training. But I really want to make sure if my training procedure has come to the plateau state, so it would be beneficial if I could compare your learning curves. Would you kindly provide the learning curves(loss vs epochs, and test psnr (on DIV2K validation) vs epochs, which are generated automatically by the released code) for training your RCAN model? If not all, x2 BI is enough for me. Thank you very much! :)
Hi, I usually didn’t use DIV2K validation images to validate in the whole training process, as they would consume much more time. Instead, I use Set5.
So the learning curve you asked is currently not available. I may upload one in the future.
Thank you for your reply. Since I only need to know whether my network has plateaued, a learning curve on set5 is also useful for me. Then is it poosible for me to have your learning curve on set5? Or can you mail the curve image to me at [email protected] if you think the learning curve shall not be seen as an official reference at this repository? I only hope to have some kind of reference, and just thank you.
@mati1994 @yulunzhang ，are you sure that the model should be trained for 1000 epochs ? I had train this model for 200 epochs, but it already costs a few days. My devices is 1080ti. Besides, if I train the model for 1000 epochs, could I get the model whose performance is the same as the paper's? Or you have some different training methods?
