Sovit Ranjan Rath

@uysalfurkan Looks like CUDA issue. Which GPU do you have, is it an RTX or GTX GPU?

@uysalfurkan Ok. I understand the issue now. Looks like `pip install -r requirements.txt` is installing torch 1.13.1 which has issues with CUDA on Kaggle. This is because of this line...

@uysalfurkan Are you running on Kaggle?

@uysalfurkan I was also facing the issue yesterday. But never saw it before. Still meed to debug. Can you try `--workers 2` and train again?

@Codehackerone I have not provided the functionality as of now. But may do it eventually.

Yes, if you can modify the code, that can work.

Sure, if you can create a clean PR, I will be happy to merge it. I made some updates to the training and datasets file. Please make sure to pull...

That will be saved in the `outputs` directory after the training completes.

Actually, I don't have a model trained right now. Its an old project and I changed my system recently, so don't have the trained model with me.

@KMCzajkowski The `add_gaussian_blur.py` script was used to generate the training data. But the `add_gaussian_blur_test_image.py` is used to generate blurry images for inference. So, it can be any value of our...