Stefan Webb

Change `min` to `tf.minimum`

You can also change the final condition with a `tf.cond`

@guyko81 the `.inv` method swaps the forward and inverse operations and returns a new transform, but it keeps the same parameters. Have you tried called `transform._inverse(Y_train)` on the original transform?

Hi @mfederici, thanks for your interest in the normalizing flow sublibrary! :) Yes, this is a bug we're aware of and something that needs to be fixed. I've moved NF...

@francois-rozet could you please file an issue on FlowTorch (separate from the discussion thread there)? I'm happy to switch over to the `softplus` solution if that is what's used in...

Thanks for picking up on this! It's an easy fix although with the developments at Meta this morning, I may need to relocate this repo to another place before I...

Hi @bigmb, thanks for submitting this issue! I don't believe we've tested on GPUs yet - I can look into this if it's failing. Could you please submit a full...