Core model implementations, experiment scripts and recorded result summaries.
results_v3/— experiment code and selected artifactsresults/— experiment code and selected artifactsscripts/— experiment code and selected artifactsresults_e2/— experiment code and selected artifactssrc/— core implementation
Inspect the relevant script’s module docstring and argument parser before running it. Experiments may require project-specific datasets, model weights and hardware. Recreate those dependencies locally; this release does not bundle them.
The model implementation starts in src/cedar_model.py and src/cedar_v2_model.py; experiment orchestration is in scripts/.
This is a research-code snapshot. Model weights, raw datasets, private operational notes and manuscript files are not included. Recorded results are provided as artifacts, not a claim of independent reproduction or production readiness.
Project artwork was generated with AI for repository presentation.
