Download the datasets:
1. NYUv2: https://www.dropbox.com/sh/86nssgwm6hm3vkb/AACrnUQ4GxpdrBbLjb6n-mWNa?dl=0
2. Cityscapes: https://www.dropbox.com/sh/gaw6vh6qusoyms6/AADwWi0Tp3E3M4B2xzeGlsEna?dl=0

Download the pretrained weights:
1. Download weights (HRNet-W18-C-Small-v2) from https://github.com/HRNet/HRNet-Image-Classification.
2. Save them to models\pretrained_models\hrnet_w18_small_model_v2.pth.

Training the Models:
1. Set the db_root (dataroot) in configs/mypath.py to where you stored the dataset.
2. Set the --storage_root and --config in train.sh.
3. Run train.sh.