search

google-research / semivl

[ECCV'24] Official Implementation of SemiVL: Semi-Supervised Semantic Segmentation with Vision-Language Guidance
Apache License 2.0
103 stars 9 forks source link

Could not find a version that satisfies the requirement #9

Closed SaSa-123 closed 2 months ago

SaSa-123 commented 7 months ago

Thanks to the author for the wonderful work. I am a beginner. I installed your requirements and created a virtual environment with python version 3.7.13. When I ran pip install -r requirements.txt, I was prompted ERROR: Could not find a version that satisfies the requirement, how to solve it? Snipaste_2024-02-20_15-33-10

lhoyer commented 7 months ago

Thank you for your interest in our work. It looks like your pip is outdated and does not know about the newer numpy version. I would suggest to upgrade pip and try it again.

SaSa-123 commented 7 months ago

Thank you for your interest in our work. It looks like your pip is outdated and does not know about the newer numpy version. I would suggest to upgrade pip and try it again.

Thank you for your sincere reply! In addition, I would also like to ask how much memory is required to run this code on the cityscapes dataset? Snipaste_2024-02-20_22-42-00

lhoyer commented 7 months ago

On Cityscapes, we trained the model with 8 x A100 (40 GB) GPUs, where each GPU has a batch size of 1 (see experiments.py at exp_id == 44.