An open API service providing repository metadata for many open source software ecosystems.

GitHub / mohamedkhayat 2 Repositories

Tunisian engineering student passionate about AI, currently diving into deep learning and applying it to projects. Exploring Java to broaden my skillset!

mohamedkhayat/fruit_internship

This internship project centers on training Vision Transformer models (RT-DETRv2) for fruit object detection using YOLO-format data. The pipeline is fully modular, supports mixed precision and W&B logging, and is built with scalability in mind. It will be extended into multi-task learning to predict bounding boxes, weight, and volume simultaneously

Language: Python - Size: 1.83 MB - Last synced at: 2 days ago - Pushed at: 3 days ago - Stars: 0 - Forks: 0

mohamedkhayat/modern-training-pipeline

This project is a modern, modular, and production-ready deep learning training pipeline built with PyTorch. It’s designed for image classification tasks and supports advanced tools like Hydra for config management, Weights & Biases for experiment tracking, and Albumentations for rich augmentations.

Language: Python - Size: 51.8 KB - Last synced at: about 1 month ago - Pushed at: about 1 month ago - Stars: 0 - Forks: 0

mohamedkhayat/DIYNeuralNet

A lightweight deep learning framework implemented from scratch using NumPy/CuPy. supports customizable architectures, forward and back propagation, dropout, He/Glorot init, and mini-batch training. Designed for flexibility,it provides a foundation for building neural networks while giving insights into the inner workings of deep learning models

Language: Python - Size: 21.2 MB - Last synced at: about 1 month ago - Pushed at: about 1 month ago - Stars: 3 - Forks: 0

mohamedkhayat/face-auth-pipeline

A real-time face verification system built with PyTorch and FastAPI, featuring a custom Hybrid Triplet Loss with margin scheduling, ResNet-50 backbone, MTCNN detection, and W&B integration for training insights. Supports image embedding extraction and API-based face comparison.

Language: Python - Size: 37.1 KB - Last synced at: about 2 months ago - Pushed at: about 2 months ago - Stars: 1 - Forks: 0

mohamedkhayat/DQN_Snake

This repository hosts a Deep Q-Network (DQN) agent designed to play the classic Snake game. The project leverages PyTorch for the deep learning component and Pygame for game rendering. This initiative served as my introduction to the fascinating world of reinforcement learning.

Language: Python - Size: 2.07 MB - Last synced at: about 2 months ago - Pushed at: about 2 months ago - Stars: 3 - Forks: 0

mohamedkhayat/spam-detection

A spam detection app using TensorFlow for model training and Flask for deployment. Built as a learning project, it applies NLP techniques from the Machine Learning Specialization and Natural Language Processing in Action, with a Java frontend for user interaction.

Language: Python - Size: 22.9 MB - Last synced at: 11 months ago - Pushed at: 11 months ago - Stars: 0 - Forks: 0

mohamedkhayat/ButterflyClassifier

This project lets users classify butterfly images using a JavaFX GUI for file selection. Images are sent to a Python backend where a fine-tuned VGG model analyzes them. The classification results are then returned to the Java application for display.

Language: Python - Size: 19 MB - Last synced at: 12 months ago - Pushed at: 12 months ago - Stars: 0 - Forks: 0

mohamedkhayat/Iris-ANN-From-Scratch

This repository contains an ANN built from scratch using NumPy and Python to classify the Iris dataset. It demonstrates the fundamentals of constructing and training an ANN without high-level libraries like TensorFlow or PyTorch.

Language: Java - Size: 12 MB - Last synced at: 12 months ago - Pushed at: 12 months ago - Stars: 1 - Forks: 0

mohamedkhayat/mnist-digit-recognition

This project is a digit recognition system using a neural network model trained on the MNIST dataset. Users can draw digits in a PyGame window, and the model will recognize the drawn digit. This project was created as a learning exercise to apply concepts from an intro to deep learning course and to incorporate an interactive element to the models

Language: Python - Size: 25.1 MB - Last synced at: 12 months ago - Pushed at: 12 months ago - Stars: 0 - Forks: 0