VicenteVivan / geo-clip

This is an official PyTorch implementation of our NeurIPS 2023 paper "GeoCLIP: Clip-Inspired Alignment between Locations and Images for Effective Worldwide Geo-localization"
https://arxiv.org/abs/2309.16020
MIT License
90 stars 13 forks source link

Loss function used for training #3

Closed t0tl closed 3 months ago

t0tl commented 4 months ago

Hello, me and a partner are interested in finetuning the GeoCLIP model however we are unsure of the implementation of the loss function. Could you share the loss function you used or give any tips for implementing it?

t0tl commented 3 months ago

In #6 the authors answer this issue. Fyi, you will still have to fix the code yourself as the queue implementation is broken in the repo.

VicenteVivan commented 3 months ago

Hi Timothy,

Thank you for your patience and for pointing out the problem. We just updated the repository and corrected the training loop as well as the queue implementation. Apologies for any inconvenience this may have caused. If you have any other questions regarding the implementation please let us know, we would be happy to help.

Sincerely, Vicente