Skip to content

Add README to APG experiments folder #1145

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
anwai98 merged 1 commit into from
Jan 24, 2026
Merged

Add README to APG experiments folder #1145

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
22 changes: 21 additions & 1 deletion scripts/apg_experiments/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,3 +1,23 @@
## Experiments for Automatic Prompt Generation (APG)

More explanation coming soon!
This folder contains evaluaton code for applying the new APG method, built on `micro-sam` to microscopy data using the `micro_sam` library. This code was used for our experiments in preparation of the [manuscript](https://openreview.net/forum?id=xFO3DFZN45).

Please note that this folder may become outdated due to changes in function signatures, etc., and often does not use the functionality that we recommend to users. We also will not actively maintain the code here. Please refer to the [example notebooks](https://github.com/computational-cell-analytics/micro-sam/tree/master/notebooks) and [example scripts](https://github.com/computational-cell-analytics/micro-sam/tree/master/examples) for well maintained and documented `micro-sam`'s APG examples.

### Evaluation Scripts:

The top-level folder contains scripts to evaluate other models with `micro-sam`, and the `plotting` subfolder contains scripts for visualizations and plots for the manuscript.

- `analyze_posthoc.py`: Experiments to visually understand and debug the APG method.
- `perform_posthoc.py`: Experiments to store results to work with `analyze_posthoc.py`.
- `prepare_baselines.py`: Experiments to run all methods presented in the manuscript with default parameters on all microscopy imaging data.
- `run_evaluation.py`: Experiments related to APG for understanding the hyperparameters using grid-search.
- `submit_evaluation.py`: Scripts for submitting jobs to slurm.
- `util,py`: Scripts containing data processing scripts and other miscellanous stuff.
- `plotting`/
- `calculate_mean.py`: Scripts to calculate mean performance metric over per modality per method.
- `plot_ablation.py`: Scripts to show results for comparing connected components- vs. boundary distance-based prompt extraction.
- `plot_average.py`: Same as `calculate_mean.py`. The scripts plot the mean values in a barplot and displays absolute rank per method over all datasets per modality.
- `plot_qualitative.py`: Scripts to display qualitative results over all datasets.
- `plot_quantitative.py`: Scripts to display quantitative results over all datasets.
- `plot_util.py`: Stores related information helpful for plotting.