Quantifying Prediction Performance

Error Metrics Calculation

• Script: RMSE_MSE_calculation_code.R

Rscript RMSE_MSE_calculation_code.R \ --pred-dir results/predictions/ \ --obs-file data/input/Raw_data_table_all_farms.csv \ --out-file results/RMSE_MSE_observed_results.csv

This script reads the observed and predicted co-occurrence data from data/input/Raw_data_table_all_farms.csv. For each farm pair, it computes: Mean Squared Error (MSE) and Root Mean Squared Error (RMSE). It can also be run locally (without requiring HPC) by executing the R script directly on a workstation.

The final summary file is saved as results/RMSE_MSE_observed_results.csv.

Visualization of Observed Results Visualization of Observed Results

• Script: Plost_for_RMSE_MSE_observed_results.R

This script runs locally on your computer (no HPC needed) and generates several diagnostic plots to assess prediction performance on the non-permuted data