Label Distribution, Class Balance & Class Weights
This lesson is part of the course
In this lesson
You calculate the share of each label, the imbalance ratio, balanced class weights and the accuracy a model would get by always predicting the biggest class.
Buy the course once for $9.99 to unlock every lesson and lab, the final exam and the certificate – lifetime access, no subscription.
Lessons in this course
- Label Distribution, Class Balance & Class Weights Locked
- Percent Agreement vs Cohen's Kappa: the Rare-Label Trap Locked
- Fleiss' Kappa: Agreement with More than Two Annotators Locked
- Majority Vote, Label Confidence & Expert Review Locked
- Gold-Standard (Honeypot) Accuracy & Annotator Flags Locked
- Label Noise: What Wrong Test Labels Do to Measured Accuracy (with a Chart) Locked
- Labelling Cost & Throughput: Hours, Review Overhead & Cost per Item Locked
- Active Learning vs Random Sampling: Labels to Target & Budget Saved (with a Chart) Locked