Infotech College of Business and IT - Machine Learning with Advanced Python

Project 01 - Animal Image Classifier using CNN

Project Overview

This project focuses on building a Convolutional Neural Network (CNN) to classify images of animals. The model is developed using PyTorch and leverages transfer learning techniques.

Domain

Computer Vision

Dataset

The project utilizes the Animal Image Classification Dataset. Dataset Source: Kaggle: Animal Image Classification Dataset

Key Skills Practiced

• Image Preprocessing
• CNN Architecture with PyTorch
• Model Evaluation (Accuracy, Precision, Recall, F1-Score, and Confusion Matrix)

Tasks

The following tasks were performed as part of this project:

• Load and preprocess the image dataset
• Design a 2-3 layer CNN (using a pre-trained ResNet-18 model)
• Train, validate, and test the model
• Visualize predictions using Matplotlib

Folder Structure

The project is organized with the following directory structure:

How to Run the Code

1. Clone the Repository (if applicable):

   git clone <your-repo-url>
   cd Animal_Image_Classifier

2. Install Libraries: Ensure you have all the required libraries by running:

   pip install -r requirements.txt

   (torch, numpy, torchvision, matplotlib, time, os, PIL, tempfile, scikitlearn, seaborn, pandas)

3. Prepare the Dataset: Download the dataset from the Kaggle link provided above and place the train, val, and test subdirectories into the 1_data folder as per the folder structure.

4. Run the Jupyter Notebook: Open the your_project_notebook.ipynb (or similar name) located in the 3_notebooks directory using Jupyter Notebook or JupyterLab and execute the cells sequentially.

   The notebook will:

   • Load and preprocess images.
   • Initialize and train two models: a fine-tuned ResNet-18 model and a ResNet-18 feature extractor.
   • Save the trained models in the 2_models directory.
   • Visualize model predictions.
   • Generate and save classification reports and confusion matrices in the 4_results directory.
   • Display training and validation graphs (if plotting code is included to save them in 5_outputs/2. Graphs).

Model Performance and Evaluation

The project evaluates the performance of both the fine-tuned CNN model (model_ft) and the transfer learning feature extraction model (model_conv) on the validation dataset.

The evaluation includes:

• Loss and Accuracy per Epoch: Printed during the training process for both training and validation phases.
• Classification Report: Provides precision, recall, f1-score, and support for each class.
• Confusion Matrix: Visualizes the number of correct and incorrect predictions for each class.

These evaluation reports (_classification_report.txt and _confusion_matrix.png) are saved in the 4_results directory for both models, allowing for direct comparison of their classification abilities.