Skip to content

Pseudo-labeling, Virtual Adversarial Training (VAT) and Attention-based methods.

Notifications You must be signed in to change notification settings

shrebox/Semi-supervised-Image-Classification

Repository files navigation

Semi-supervised Image Classification

The projects aims to tackle the problem of data scarcity for training data-hungry neural networks using the semi-supervised learning which relies on small amount of labelled dataset. Three techniques namely - Pseudo-labeling, Virtual Adversarial Training (VAT) and Attention-based methods have been implemented and analyzed on CIFAR-10 and CIFAR-100 datasets.

Checkout Problem Statement and Report for more details.

About

Pseudo-labeling, Virtual Adversarial Training (VAT) and Attention-based methods.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published