Skip to content

elliotjones103/Animal_Classification

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

21 Commits
 
 
 
 

Repository files navigation

Classification Project

This project explores image classification by training a model to identify different types of animals from images. The main aim is to refresh my Python skills while building a practical machine learning project.

Overview

Through this project, I will work through the full image classification pipeline, from downloading a dataset to training and improving a model. The project will also help me strengthen my understanding of how image datasets are structured and how machine learning models learn from visual data.

Project Goals

The main goals of this project are to:

  • download and work with image datasets from Kaggle
  • explore the dataset structure and understand how image data is stored
  • clean and inspect the data before training
  • practise using Python libraries commonly used in machine learning projects
  • build a classification model to identify different animals from images
  • improve the model over time to reduce training and validation loss

Tools and Libraries

  • Python
  • pathlib
  • os
  • matplotlib
  • torchvision
  • PyTorch

Project Steps

  1. Download the dataset from Kaggle
  2. Explore the classes and image folders
  3. Inspect image samples and class balance
  4. Prepare the dataset for training
  5. Build a classification model using PyTorch
  6. Train and evaluate the model
  7. Improve performance through experimentation and tuning

Purpose

The purpose of this project is both educational and practical. It allows me to revisit core Python concepts such as file handling, loops, and working with libraries, while also developing a better understanding of machine learning and image classification.

About

Image classification project using PyTorch to build and train a CNN on animal image data.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors