Questions about reproducing experiment

[tjdgh0715]

Dear authors,

Thanks for the great work!

I tried to reproduce the main experiments, but got weird errors & symptoms.

I followed the Rainy cityscapes setup that is explained in the git instructions.

However, during the training, I consistently got these errors after the 1st epoch:
I directly followed the training command instruction as:
python run_experiments.py --exp -1 --wandb 1

FileNotFoundError: img file does not exist: ./data/Cityscapes/gtFine/val/rainy_image/munster/munster_000090_000019_gtFine_labelTrainIds.png
FileNotFoundError: img file does not exist: ./data/Cityscapes/gtFine/val/rainy_mask/munster/munster_000090_000019_gtFine_labelTrainIds.png

It seems that there are no rainy_image/ and rainy_mask/ directory after the source data preprocessing, when followed the instructions.

I also attached the error log for better understanding.

output.log

Can I get some advice to solve these error, and successfully reproduce the experiments that is conducted in the paper?