Skip to content

Project for computer science course implementing triplet loss learning

Notifications You must be signed in to change notification settings

alailink/pytorch_tripletloss

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

pytorch_tripletloss

Project for computer science course implementing triplet loss learning. Through this project I demonstrate my ability to use pytorch for image classification using AlexNet, transfer learning, from-scratch networks, and from journal descriptions. I also use industry-standard train, test, and validation data splits for accuracy reporting.

alt text

Overview

Through a series of problems, I compare the performance of several convolutional neural networks (CNNs). Throughout I use the NWPU-RESISC45 dataset (Northwest Polytechnic Institute: REmote Sensing Image Scene Classification).

In problem 1 I use a pre-trained AlexNet, which I then modify through transfer learning to increase the test accuracy by 5%.
In problem 2, I generate my own CNN in pytorch which I name ClayNet.
In problem 3, I take a journal paper which describes triplet margin loss, create the architecture in pytorch, and implement on our dataset.

alt text

About

Project for computer science course implementing triplet loss learning

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published