cavalleria / cavaface

face recognition training project(pytorch)
MIT License
456 stars 88 forks source link

Fix megaface eval as only one out is returned #13

Closed xsacha closed 4 years ago

xsacha commented 4 years ago

This was changed recently (but before this was pushed) to have only one output. So pulling first element now breaks it.

I had to make numerous changes to get this to work with Python 3 but I believe this change would break in both. I also have an issue where it finds the GPU free memory size in MB instead of in GB but haven't investigated.

cavalleria commented 4 years ago

Thanks for your pull request. I forgot to commit this bug.