Dataset Directory Structure

[bbbts]

Hi, Could you please provide the way your dataset is arranged/ structured in your directory (i.e. train data, validation data and test data). I tried running the downloading script to download the ade20k dataset but I am running into some errors. Hence, I downloaded the ade20k dataset manually from Kaggle. Please kindly help how I should arrange them in my directory that will adhere to the train.py and eval.py file in the 'scripts' folder. Let me know how I can arrange the test images as well.
Thanks!
This is how I have kept it currently.

And this is how I have edited the ade.py file in dataloader folder -