post hoc plot utils that generate summary metrics and plots

aidecentralized / sonar

SONAR - Self-Organizing Network of Aggregated Representations

MIT License

13 stars 34 forks source link

post hoc plot utils that generate summary metrics and plots #123

Closed joyce-yuan closed 4 weeks ago

joyce-yuan commented 1 month ago

Usage: First, fill out the path to the logs of your experiment log_path = your_logs_path_here

Then run python src/utils/post_hoc_plot_utils.py

So far this set of plot utils logs user metrics and summary metrics (including AUC and best test / train loss + accyracy) under aggregate_metrics, and plots test / train loss + accuracy per round under plots