rpotter6298
a721e52909
- Introduced `poster_model_comparison.py` for generating model comparison figures. - Added `plot_poster_roc_comparison.py` for creating ROC comparison figures for PAPILA binary classification. - Created new JSON configuration files for clinical solo models with and without geometry injection. - Implemented batch dispatch updates in `batch_dispatch.py` to utilize run names from configurations. - Added analysis scripts: `compare_grid.py`, `inspect_embeddings.py`, and `summarize_run.py` for evaluating model performance and feature embeddings. - Created experiment configurations for various training scenarios, including warm sweeps and promoting successful runs. - Added binary ROC comparison and model comparison figures to the results directory.
11 lines
289 B
JSON
11 lines
289 B
JSON
[
|
|
{
|
|
"_note": "Single rep of tritower default with save_features=true so we can inspect nt-stage embeddings (Hadamard-product collapse hypothesis).",
|
|
"run_name": "experiments/tri_v1/baseline_tri_features",
|
|
"reps": 1,
|
|
"overrides": {
|
|
"save_features": true
|
|
}
|
|
}
|
|
]
|