Closed fnzhan closed 3 years ago
Does it help? And how big is your dataset? Mine is 70k and I see validation loss lagging more and more behind from less than 10 epochs.
I initially thought that the 256x256 training images are randomly cropped from the dataset you provide. But that turned out to be not the case and I had to provide a set of 256x256 crops myself. I see no reason why random cropping should not work (other than slightly higher overhead) and it may help with overfitting but I did not get to the bottom of this yet.
For my part, overfitting is caused by the removal of random crop. After adding the random crop in model training, the validation loss goes well.
Could adding random crop to the training data improve my image2image model? https://github.com/CompVis/taming-transformers/issues/51 I based the code off of the imagenet code.
I have around 11k training examples.
Would there a way to do pre-training?
I also think that the stage 2 training of VQ-GAN would suffer from the overfitting on FFHQ, because this repository does not contain data augmentations for FFHQ training dataset.
I train the transformer but find it overfits after 30-40 epochs, with the validation loss goes high and the training loss is very small. If you meet this problem in model training. Now I try to use the pkeep=0.9 in the cond_transformer.py to avoid overfitting.