vidurbervilles

Hi! I found your library while working on the code from a paper about anomaly detection with VAEs. I was just wondering if you are still working on trixi? Or...

Hi! I am experimenting with variational autoencoders for anomaly detection, but I can't find the function fillup_pad (originally called FillupPadTransform). Has it been dismissed?

How to improve the reconstruction of high-frequency details in the VQVAE training?

16

Thank you so much for sharing! Could you provide insights into the number of epochs required to achieve high-resolution, fine details during VQVAE training for 256x256 RGB images? Additionally, has...

'MulticlassJaccardIndex' object has no attribute 'plot'

2

Hi there! I encountered the error mentioned in the title while experimenting with the notebook for CellPose training using the UNI encoder. Could it be that `MulticlassJaccardIndex` has been updated...

Hi PyWavelets team, I'm trying to locate the exact place in the codebase where the filter coefficients (e.g., for db3) are defined. I'm aware that they can be accessed at...

Rationale for Using 1/4 Input Resolution in Prediction Head?

1

Hi, I’m curious about the reasoning behind a particular architectural design choice I noticed. For all predictions except the refinement vector, features are extracted from the backbone at a resolution...

Questions on color augmentations and cropping during Cellpose SAM fine-tuning

1

Hi, I’m fine-tuning Cellpose SAM and have two questions: 1. **Color/intensity augmentations** The paper states that, during training, images undergo: * 25% contrast inversion (`x → 1 − x`) *...