Minchul Kim

I faced the same problem and found a solution by using DistributedDataParallel from Nvidia apex.

Hi xingjunhong Thanks for the interest but I am not sure what the question is. For discussion related to the gradient, I would like to check out this discussion, https://github.com/mk-minchul/AdaFace/issues/24...

You may use the option `--start_from_model_statedict`. Specify `start_from_model_statedict` to be the pretrained model ckpt path. And use your own image dataset as specified in [README_TRAIN.md](https://github.com/mk-minchul/AdaFace/blob/master/README_TRAIN.md)

Hi WOTMCCR This error is due to the update in pytorch lightning. You could use the `pytorch-lightning==1.4.4 ` to use the existing code. Or if you would like to use...

code is updated as well.

Thanks for the notice! I have updated the code for backward compatibility.

Hi longzewei I am not sure what you mean by the question.

Dooray! Failure Notice Failure Notice Your message sent to ***@***.*** has failed to be delivered. Please refer to the below for details. * Recipient :  ***@***.*** * Sent time : ...

Hi trnikon. The agedb30 is one of the 5 validation sets that are used for tracking the model performance during training. If you do not need these, you should change...

Dooray! Failure Notice Failure Notice Your message sent to ***@***.*** has failed to be delivered. Please refer to the below for details. * Recipient :  ***@***.*** * Sent time : ...