Sébastien Tosi

Impractical GPU memory requirements

4

While indeed extremely fast, the GPU memory requirement is impractical on my setup: about 8 GB for a 1024x1024x19 image (16-bit) and a tiny 32x32x16 PSF. For images slightly above...

Labkit incompatible with batch mode

10

Labkit seems incompatible with batch mode when calling it from a macro. I hint that it comes from the fact that the results is a virtual stack. Would it be...

I wonder why input normalization has so much impact on the training of UNets when the first batch normalization is actually performed very early in the network (and should play...

UNet Implementation details

2

I noticed some subtle differences in the implementation of your UNet and the one in this repo: https://github.com/johschmidt42/PyTorch-2D-3D-UNet-Tutorial. Basically, the batch normalization is performed after the ReLU layers instead of...

Sparse annotations for training

2

Is there currently a way to use sparse annotations for training, i.e. for pixel classification by ignoring all the pixels set to 0 in the target image during loss computation...

I'm using the last version of the plugin with the last version of Napari under Windows 10 and regardless of the settings and images I'm using, nothing is outputted to...

TIFF images

6

Are you planning to add support for TIFF images (possibly multi-page)?

Image size limit

4

It seems that there is an image size limit around 10k x 10k. For larger images no error is issued but no output is saved.

I'm struggling with the no permissive pip requirements of empanada-napari. Especially, forcing numpy==1.22 is very constraining and requiring Napari 0.4.18 makes it difficult to install this plugin with some others....

Any chance that cellpose-napari gets updated to fully support Cellpose 4? In general Napari plugins meant for an underlying BIA library/tool are poorly maintained, or their compatibility with newer versions...