mikonvergence
mikonvergence
Hi, @cpnovaes, thank you! Indeed, that would require some changes to the code examples, but the current conditional model allows to specify the `condition_channels` at initialisation (see https://github.com/mikonvergence/DiffusionFastForward/blob/master/src/PixelDiffusion.py): ```python class...
Hi @cpnovaes! That's exactly the right approach with the PixelDiffusion type. For the latent diffusion, that will be tricky, because: * The Autoencoder has been trained on natural images with...
Hi, @SkalskiP! Any progress with verification?
Hi @blumenstiel - thanks for bringing this up! I had a look too and it does seem like these two cells are indeed corrupted. We made no changes to the...
Hi @danielz02, thanks! This should be correct. The average time shift between Core-S2 and Core-S1 is actually about 7 days.
Hi @noahdasanaike! Have you tried the example filtering notebook? If yes, could you provide some more detail about the use cases that you're interested in?
Great suggestion, will try to fix this soon (hopefully along with expanded set of notes)
Yes, I've not had a chance to review the remaining notes due to other commitments, so they were never added to the official project. It seems that there is some...
Update (copied from the other issue): I finally had time to look at what we put in the paper. The ESA original implementation is in my view closer to that...
Great suggestions! I'll try to prioritise and get it fixed up soon along with the several identified bugs. At the moment, I think the last functionality (find all cells that...