Adam Tyson
Adam Tyson
Lazy loading of directories of images seems to be in core napari, so we can do without this plugin. (also remove from other repos).
@dstansby Nothing has changed with cellfinder, is this something to do with the actions?
Linked to #64. Allowing upper layers to be frozen when retraining should make slight tweaks easier
Although the background channel is useful for the cell classification and registration steps, it is not always practical to double the data collected (and in the case of some fluorophores,...
On tagged commits, rather than test then deploy, the workflow should be: Build wheels -> Install from wheel -> Test This should catch issues building the package, in the manifest...
Something users can copy and paste when writing up papers.
There are a number of users who have 3D data, but require 2D cell detection (i.e. low Z sampling). Theoretically this should be possible to add within the cellfinder framework....
Users are generating small amounts of data, and the network seems to be over fitting. They're having to manually get existing data to constrain their models. This could be improved...
The default batch size for classification (32) was chosen so it would run on pretty much any good GPU but as @larsrollik has found, this can be increased a lot,...