Skip to content

Latest commit

 

History

History
10 lines (6 loc) · 452 Bytes

README.md

File metadata and controls

10 lines (6 loc) · 452 Bytes

Solution repository for Infineon Machine Learning Hackathon

Team Name : Idk

Private Leaderboard National Rank : 3 (Accuracy = 0.7485)

Task : Document Tagging Problem

Methodology : This is a multi-class classification system aimed at identifying a subset of 46 classes to which a particular piece of text may belong. Utilised the technique of layer freezing, and subsequently implemented it for achieving an accuracy of 74.85% on Infineon dataset.