DECODE
DECODE copied to clipboard
TuragaLab
This is the official implementation of our publication "Deep learning enables fast and dense single-molecule localization with high accuracy" (Nature Methods)
### Discussed in https://github.com/TuragaLab/DECODE/discussions/229 Originally posted by **Abdullah12Husain** February 21, 2024 Hello, I have been using DECODE for the last year or so and it has always worked smoothly in...
SpatialIntegration: Two emitters with prob tensor = [0.95, 0.98] end up as [1., 1.] which they shouldn't. Edge case.
Decouple ROI to Frame from SplinePSF. General ROI to Frame should: - Map stack of rois to stack of frames ROI: - size N x H x W Frame: -...
version: 0.10.0 e.g. on colab Solution: - On the fly convert new style .mat or change loader to allow for newer style
Ideas: - parse what's after the hashtag for a value - establish separate .yaml file with same structure but tooltip at as value
Change all nested lists to not nested ones, i.e. `psf_extent: [[-0.5, 39.5], [-0.5, 39.5], None]` to ``` psf_extent: x: [-0.5, 39.5] y: [-0.5, 39.5] z: None ``` Reason: Makes it...
Add instructions and reflect them in the .yaml parameter file for 2D fitting.
Should be settable by cmd line flag, implementation something `logging.root.addHandler = RotatingFileHandler`
Save model during training as self-contained binary to make the model less sensitive to version changes. Test thoroughly.
Relative paths to __file__ or similiar could likely fail in packaged decode installations (because then folders don't exist anymore?). Do it similarly to load_reference, i.e. ```python try: import importlib.resources as...