training on cityscapes dataset

ZiqinZhou66 / ZegCLIP

Official implement of CVPR2023 ZegCLIP: Towards Adapting CLIP for Zero-shot Semantic Segmentation

MIT License

195 stars 17 forks source link

Open chrisyan opened 5 months ago

chrisyan commented 5 months ago

Dear authors,

thank you for your amazing work! Could you guide me how to train on the cityscapes dataset?