Open Huang-yihao opened 3 years ago
You can use the following PyTorch PyNET code, it requires just a few modifications at the top level.
I have trained on EBB dataset with Pytorch PyNET. Details: Training on level2, psnr on valid set(sampled some pic from ebb) can reach ~23. But on level1, psnr on same valid set only ~4. The result on visual set all white. Difference: I checked difference with tf-version, only find tf-version no instance normalization at level 1. So what's difference with tf-version, besides
I have trained on EBB dataset with Pytorch PyNET. Details: Training on level2, psnr on valid set(sampled some pic from ebb) can reach ~23. But on level1, psnr on same valid set only ~4. The result on visual set all white. Difference: I checked difference with tf-version, only find tf-version no instance normalization at level 1. So what's difference with tf-version, besides
Do you have EBB dataset link? The official link is not provided now.
Will you provide Pytorch version of the code and model in the future?