How to valuate during traing

facebookresearch / moco

PyTorch implementation of MoCo: https://arxiv.org/abs/1911.05722

MIT License

4.73k stars 777 forks source link

Closed zjcs closed 4 years ago

zjcs commented 4 years ago

I notice the finnal R@1 after 200epoches in ReadMe is 60%, but there is no code to valuate the model in the repo, only training accuracy.

Can you help me that how to valuate the perfomance during the training?

KaimingHe commented 4 years ago

I am not aware of a reliable way of validating during training. You may consider the kNN classifier adopted in the InstDisc paper.