This repository contains a deep learning project that implements image classification on the CIFAR-10 dataset using the Res2Net model. The project demonstrates how to set up and train a Res2Net model for image classification tasks and evaluates its performance.

Project Overview This project follows these steps:

Model Selection: Uses the Res2Net model architecture. Data Preparation: Processes and prepares the CIFAR-10 dataset. Model Training: Trains the Res2Net model on the CIFAR-10 dataset. Evaluation: Computes and reports the accuracy of the trained model.

Requirements Python 3.x PyTorch torchvision timm

Install the required packages using:

pip install torch torchvision timm

Dataset The CIFAR-10 dataset is used for training and evaluation. It consists of 60,000 32x32 color images in 10 classes, with 6,000 images per class. https://www.cs.toronto.edu/~kriz/cifar.html

Implementation Details Model: The Res2Net model from the timm library is used, pre-trained on ImageNet. The output layer is adjusted for 10 classes to match the CIFAR-10 dataset. Data Transformation: Images are resized and normalized according to standard practices for the Res2Net model. Training: The model is trained for 5 epochs using the Adam optimizer with a learning rate of 0.001. Evaluation: Accuracy is computed on the test dataset after each epoch.

Results During training, the test accuracy of the model will be printed after each epoch.