Armand Collin
Armand Collin
**Describe the problem** The current inference pipeline for nnUNetv2 models is located at https://github.com/axondeepseg/nn-axondeepseg There are some minor issues (i.e. color profile conversion sometimes required, see https://github.com/axondeepseg/nn-axondeepseg/issues/5) that would require...
### Discussed in https://github.com/axondeepseg/axondeepseg/discussions/766 Originally posted by **ktyssowski** October 26, 2023 Axondeepseg is working great for me for separating myelin from axon, but I think the instance segmentation that happens...
**Describe the problem** When you load an image in napari and click on `Apply ADS model`, the user is prompted for the pixel size of the image. However, after this...
## Checklist - [x] I've given this PR a concise, self-descriptive, and meaningful title - [x] I've linked relevant issues in the PR body - [x] I've applied [the relevant...
I was playing around with the `download_models` script and its behavior is not really ideal. For example, I found that #394 was still a problem (PR fixing it was closed...
In the wiki, we have a few pages that are either very outdated (https://github.com/axondeepseg/axondeepseg/wiki/Datasets) or downright useless (https://github.com/axondeepseg/axondeepseg/wiki/How-to-manage-the-Binder-link what even is this?). This wiki may confuse people who will work...
In order to support 3-class segmentation (context: unmyelinated fibers), we will need to change some stuff on the ADS side. For ivadomed, nothing really changes: we will use 3 ground-truths...
## Checklist - [x] I've given this PR a concise, self-descriptive, and meaningful title - [x] I've linked relevant issues in the PR body - [x] I've applied [the relevant...
**Describe the problem** Sometimes, we suggest users to try other models available under the axondeepseg org. There is no "centralized" place where all models are described, and if the user...
**Describe the problem** The _download_model.py_ script currently takes models by downloading the whole repository and extracting the model folder. It seems we used to put the `.pt` checkpoint files directly...