Skip to content

Experiments to quantify how an imbalanced or few-shot dataset will impact a model (SVM)

Notifications You must be signed in to change notification settings

Tyler-Hilbert/FewShot_SVM_Question

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

37 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Purpose

This repo contains a couple different experiments to quantify how an imbalanced or few-shot dataset will impact a model.
The purpose of this experiement is to quantify the impact of the imbalance. This isn't designed to be a SOTA model for solving the imbalance.

Dataset

The dataset used for this experiment is a modified verion of MNIST where the minority class is limited to either 3 or 50 training examples.

About

Experiments to quantify how an imbalanced or few-shot dataset will impact a model (SVM)

Topics

Resources

Stars

Watchers

Forks

Languages