microsoft / metric-transfer.pytorch

Deep Metric Transfer for Label Propagation with Limited Annotated Data
MIT License
49 stars 19 forks source link

Supervised Metric Pretraining #1

Open chingisooinar-second opened 3 years ago

chingisooinar-second commented 3 years ago

Hello, I loved reading your paper. However, I couldn't find a Supervised Metric Pretraining. Can I contribute by adding it? Thank you

bl0 commented 3 years ago

Hi, thanks a lot for your interest in our work. May I ask which supervised metric pre-training method you want to add?

chingisooinar-second commented 3 years ago

I am currently working with a labelled dataset having direct commonalities with my target task. In the paper, you mention an equation using which we can maximize the likelihood of each example being supported by other examples belonging to the same category. I couldn't find a sample code in the repository, so I was wondering if I can add that.

bl0 commented 3 years ago

Thanks. You can try to submit a PR and I will check that.