Fediory / HVI-CIDNet

"You Only Need One Color Space: An Efficient Network for Low-light Image Enhancement"
MIT License
125 stars 15 forks source link

how to use pretrained for custom dataset #34

Open trongldwork opened 6 days ago

trongldwork commented 6 days ago

i have a yolo image datasets, i want to inhance it not by training, how to prepare it and inference

Fediory commented 4 days ago

You can eval in the Custome Datasets ways us follows python eval.py --unpaired --custome --custome_path ./your/costome/dataset/path --unpaired_weights ./weights/LOLv2_syn/w_perc.pth --alpha 0.9

alpha is a light adjust weight. Higher is lighter unpaired_weights is the path of our model weights, which you can choose what you want (detailed in Readme). Change custome_path to your YOLO path.