Skip to content

Oyebamiji-Micheal/T-Shirt-Sleeve-Classification-using-Convolutional-Neural-Network

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

T-Shirt Sleeve Classification using CNN

Language ML Tool Framework reposize Topic Model

A beginner's approach to deep learning with PyTorch


Table of Contents

Overview

This project is a beginner's approach to deep learning with PyTorch - so nothing too fancy like transfer learning 😐, regularization 😑 or pre-trained models 🙁. The project aims to classify the sleeves of T-shirts into three categories: full sleeve, half sleeve, and sleeveless.

Dataset

The dataset I used was obtained from Kaggle.

It is categorized in 3 classes, each representing a different type of sleeve:

  • Full Sleeve: 2118 images
  • Half Sleeve: 2670 images
  • Sleeveless: 1537 images

Model

The model used in this project consists of three convolutional blocks, each containing two convolutional layers followed by ReLU activation and max pooling. The details of this architecture can be found in the notebook.

Result

After training the model, I evaluated its performance on the test set. The model achieved 93% accuracy and 93% F1-score on the test set, which is hmmm somehow impressive.

However, the model does have a hard time differentiating between half sleeve and sleeveless cloths. Not surprising though. What I will try to do next is use some regularization techniques and maybe a pre-trained model.

See you in the next one 🙂

About

A beginner's approach to deep learning with PyTorch

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published