Maitreyapatel / reliability-checklist

NLP tool for wide-range model reliability evaluations

Home Page:https://reliability-checklist.readthedocs.io

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

feature: radar chart for better comparison of different models

Maitreyapatel opened this issue · comments

Is your feature request related to a problem? Please describe.
Currently, reliability-score stores the CSV file having all the results. It would be beneficial to have the radar plot to efficiently compare the results between multiple models into a single plot.

Describe the solution you'd like
A callback that creates the radar chart and also takes the other model's reliability results as input for comparisons.

Describe alternatives you've considered
wandb #19 can allow the comparisons between models but it not efficient as everyone needs to consider various metrics to make a single decision.

Additional context
N/A