Jed Homer
Jed Homer
Thanks for your reply and the recommendations! The main thing I am looking to do is a Haar 3D-DWT as part of a parallelised layer in `Pytorch`. I think all...
No I didn't - it's an area of mystery to me but I would be very interested in a discussion / looking into a solution! There doesn't seem to be...
Following @rosinality's advice, it works with label conditioning on MNIST. I added a label conditioning to the affine coupling layers instead of the prior (this is good because the net...
Thanks @patrick-kidger, I will definitely use `equinox` next time around, its design is incredible.
Hi Johanna! Thank you for the reply! I might be missing something here... If I change the batch size e.g. `x = jnp.ones((7, 4))` the error returns. It seems like...
@lockwo - thank you for this insight! This was what I was trying just now (didn't work out the tree mapping, been confused by tracers and what not...). If it's...
Awesome stuff, I was hoping you'd see this @patrick-kidger :). Yeah I tried the param `xs` arg for `lax.scan` but I must have had a hidden bug or something in...
Ah this is interesting. I will have a look at this but not sure a priori!