Question about how to obtain acuity_results.csv and filtered_final_dataset.csv

[imJiawen]

Hello,

Thank you for releasing the ER-Reason benchmark and the accompanying codebase.

While trying to use the data for evaluation, I noticed that some scripts depend on CSV files that do not appear to be included in the repo, e.g.,

• Acuity/... scripts load acuity_results.csv
• Diagnosis/Llama/llama_diagnosis.py loads filtered_final_dataset.csv

I could not find these files in the repository or clear instructions on how to generate them from the PhysioNet ER-Reason dataset.

Could you please clarify how these two files are generated? Is there a preprocessing script to create them from the PhysioNet release (e.g., er_reason.csv)?

Any guidance would be greatly appreciated.

Thank you!