Deep Fake Detection Challenge

$500K in AWS Promotional Credits via AWS Promotional Credit Request Form, available at kaggle forum (1st application deadline 23 Dec):

Credits have now been distributed :smile: check your accounts!

How much credits did you get out of curiosity?

500 :slight_smile:

1 Like

I would also like to join but I’m based out of India.

Now also gcp is offering TPU for this comp https://www.kaggle.com/c/deepfake-detection-challenge/discussion/125313

Looking for a Team Mate/Team, I have already participated in the competition and have a model trained on a subset of data and have inference which runs successfully against the public validation set.

I come from cloud Infrastructure engineering background, so good with setting up and managing cloud infrastructure, automated pipelines to pre process the data for training etc.

@NathanHub Would be happy to teamup, if you are still looking for team mates.
see above for details, thanks!

Has anyone here used fast.ai to take into account the temporal nature of video by using RNN’s/LSTM’s? It seems like it would not be straightforward and might be easier to do in native pytorch instead of shoehorning it into fast.ai. Any thoughts?