search

Rudrabha / LipGAN

This repository contains the codes for LipGAN. LipGAN was published as a part of the paper titled "Towards Automatic Face-to-Face Translation".
http://cvit.iiit.ac.in/research/projects/cvit-projects/facetoface-translation
MIT License
578 stars 122 forks source link

epoch size during training #32

Closed dlwogns0128 closed 4 years ago

dlwogns0128 commented 4 years ago

In train.py or train_unet.py, the default epoch size is 20000000. But, you trained model 20 epochs with LRS2 dataset in the paper. How many epochs do I have to train the model if I want to get accuracy like your proposed trained weights?

Thanks!

prajwalkr commented 4 years ago

Please train for around 20 epochs on LRS2. We have set that so that we can just stop the training whenever desired (as convergence will be different for different datasets).