mbhenaff / een Goto Github PK

Why did you change the flow chart？Could you please explain that?
thanks.

Hi, in the paper it is said that the trained model can predict 1 or 4 future frames conditioned on previous 4 frames.
How is this achieved for predicting 1 or 4 frames? Like, what's the value of x, the input frame, and z, the latent, for predicting each frame?

What is ncond and npred??

I trained the model, but it is too slow, and I wish to use the original baseline model to check whether the training step is right? Where can I download the model=baseline-3layer-loss=l2-ncond=1-npred=1-nf=64-lrt=0.0005.model?