nkanungo / era_s13 Goto Github PK

Yolo assignment

This repository contains code related to the YOLO (You Only Look Once) assignment. The code is designed to train and evaluate a YOLOv3 model for object detection tasks.

Table of Contents

• Introduction

• Getting Started

• Usage

• Configuration

• Results

• Contributing

• License

Introduction

The YOLO (You Only Look Once) algorithm is a real-time object detection system that can detect multiple objects within an image in a single pass. This repository contains code for training and evaluating a YOLOv3 model using the PyTorch framework.

Getting Started

Clone this repository to your local machine: bash Copy code git clone https://github.com/nkanungo/ERA_S13.git cd yoloassignment Install the required dependencies. You might want to consider setting up a virtual environment before installing the dependencies: bash Copy code pip install -r requirements.txt Download the dataset and update the dataset paths in the configuration file (config.py).

Usage

Training To train the YOLOv3 model, you can execute the ipynb

This will start the training process. The training progress will be displayed, and checkpoints will be saved.

Evaluation

After training, you can evaluate the trained model using the evaluation dataset. In this code I have used training loader , however you can choose your own

Here is the link to the application

https://huggingface.co/spaces/nkanungo/yolov3_object_detection

Configuration

Configuration settings for the YOLO model and training process can be found in the config.py file. You can modify these settings according to your requirements, such as changing the number of epochs, learning rate, and anchor sizes.

Results

After training and evaluation, the results can be visualized using the metrics and plots generated during the process. You can find evaluation metrics in the logs/ directory and visualize them using tools like pandas and seaborn.

Contributing

Contributions to this repository are welcome. If you find any issues or have suggestions for improvements, feel free to open an issue or submit a pull request.