YOUSSEF MOURAFFA's Projects
This program implements a camera calibration algorithm based on Zhang's method. It allows you to calibrate your camera, estimate its intrinsic and extrinsic parameters, and correct for lens distortion effects.
A comprehensive Jupyter notebook project that uses Support Vector Machines (SVM) for the classification of breast tumors into malignant or benign categories. The notebook includes data exploration, visualization, model training, and evaluation, providing insights into breast cancer diagnosis using machine learning.
This project demonstrates image classification using the CIFAR10 dataset in TensorFlow. It compares the performance of a simple Artificial Neural Network (ANN) and a Convolutional Neural Network (CNN) for this task. The purpose is to understand why CNNs are preferred over ANNs for image classification.
Welcome to the Data Scientist Daily Helper Functions repository! π This collection houses Python helper functions simplifying common tasks in data scienceβimage processing, natural language processing, machine learning, and general programming. Boost your productivity with this versatile toolkit.
Welcome to my digital resume! I'm Youssef Mouraffa. This Streamlit-powered platform showcases my expertise in data science, machine learning, and computer vision. Dive in to explore my career journey and skill set!
I'm so glad to share my first OpenCV project using Arduino, It's a simple program for detecting faces and turning on a LED. If there is a detection then the green LED is on otherwise, the yellow LED lights up
Embark on a creative journey with AnimeGAN, a project harnessing Generative Adversarial Networks (GANs) to generate captivating anime-style images. Witness the dynamic interplay of a generator and discriminator, resulting in the creation of high-quality, indistinguishable-from-real anime art. Explore the artistry of AI in this captivating endeavor!
A neural network-based approach for handwritten digit classification using the MNIST dataset. Explore different models, techniques, and architectures to achieve accurate digit recognition.
This repository features a Python K-Nearest Neighbors (KNN) implementation from scratch, with an explanatory notebook. See KNN in action on the Iris dataset and explore data visually in a dedicated notebook. A hands-on guide to understanding and applying KNN!
This repository contains a simple implementation of linear regression from scratch in C++ and includes Python scripts for data generation and error visualization.
A Streamlit web application powered by the Gemini API for question/chat and image generation.
This project is an implementation of a Perceptron with one hidden layer and softmax function. The purpose of this project is to build a neural network that can classify input data into different categories
An educational deep dive into machine learning, building a polynomial regression model from the ground up with Python. This repo offers hands-on experience with the fundamentals of regression analysis, gradient descent optimization, and data visualization.
Comprehensive object detection using YOLOv5, trained from scratch. Includes data preparation, YOLOv5 training on 20 labels, and testing on images/videos. Utilizes Google Colab's V100 GPU for robust detection.
An advanced, Python-based sentiment analysis tool utilizing logistic regression to accurately classify Twitter messages. This project demonstrates sophisticated NLP techniques and model optimization for real-world social media data analysis.
Skimlit: NLP Model for Sentence Classification. Replicates 'Neural Networks for Joint Sentence Classification in Medical Paper Abstracts.' Using PubMed 20k RCT dataset, this project trains a deep learning model to categorize sentences in paper abstracts for efficient literature skimming by researchers.
This code generates a disparity map using the StereoSGBM algorithm. The disparity map represents the difference in pixel coordinates between corresponding points in a pair of stereo images. It provides depth information for each pixel, which can be used for tasks such as 3D reconstruction, object detection, and scene understanding.
The StereoVision-DepthEstimation project utilizes computer vision techniques to estimate real-time depth from a webcam. It employs stereo vision and facial feature tracking, providing accurate depth measurements. Explore the world of 3D perception and unlock new possibilities in computer vision.