Project Repository

This repository contains several machine learning projects that I worked on as part of a Deep Learning course.

Perceptron

This was my first introduction to machine learning, focusing on a single neuron model. The weights are adjusted using stochastic gradient descent. You can apply this model to the Fashion-MNIST dataset.

Pascal Dataset Classification

In this project, Convolutional Neural Networks (CNNs) are used to classify images from the Pascal dataset.

• AlexNet: Utilizing a pretrained AlexNet model (AlexNet), we fine-tune it to classify images in the Pascal dataset. See the implementation here.

• Custom CNN: We designed a custom CNN to classify images in the Pascal dataset. Training this network requires significant computational resources, so I used Google Cloud Platform (GCP) instances. Training on a personal computer may take considerably longer. The implementation can be found here.

YOLO Loss Function Project

This repository contains a project where I implemented the YOLO (You Only Look Once) loss function (Redmon et al. 2016) for object detection tasks.

Overview

YOLO is a state-of-the-art object detection system known for its speed and accuracy. This project focuses on implementing the YOLO loss function and applying it to the Pascal dataset.

Results

The repository includes sample results obtained from training the YOLO model using the implemented loss function in the Yolo Notebook.

References

Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. arXiv preprint arXiv:1506.02640. Retrieved from https://arxiv.org/abs/1506.02640

Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E. (2017-05-24). "ImageNet classification with deep convolutional neural networks" (PDF). Communications of the ACM. 60 (6): 84–90. doi:10.1145/3065386. ISSN 0001-0782. S2CID 195908774.