MuhammadShavaiz
Hi! I do Machine Learning, Deep Learning, and explore the Software Development Lifecycle. I build web projects that combine data science with web tech.
Lahore
Pinned Repositories
Advanced-Data-Structues-FIFA_Dataset
This C++ project showcases efficient data management for FIFA World Cup data from 1930 to 2018. It uses AVL Trees for fast searching, Hash Tables for quick access, and Singly Linked Lists for easy traversal, optimizing both memory usage and retrieval speed.
Aeon-Sentinels-Unity-Game
A Unity-based tower defense game using C#, ShaderLab, and HLSL. Features dynamic difficulty scaling and single-player mode, utilizing Unity’s physics and rendering systems.
AI_learning
Google Colab notebooks showcasing PyTorch implementations and experiments. Covers deep learning techniques, including neural networks and NLP concepts.
Clothes-Generator-GAN-s
The Clothes-Generator project uses TensorFlow and GANs to create synthetic clothing images from the Fashion MNIST dataset. Leveraging TensorFlow’s framework, the project explores advanced image synthesis techniques to improve generated image quality.
DL-Visual-Question-Answering
The Visual Question Answering (VQA) project features a model with a simple GUI that handles both images and videos. It uses OpenAI's CLIP for encoding images and questions and GPT-2 for decoding embeddings to answer questions based on the VQA Version 2 dataset, which includes 265,016 images with multiple questions and answers.
Face-Generation-Wasserstein-Gan-s-GP
This project demonstrates a GAN built with PyTorch, using a subset of 5000 CelebA images. It leverages Wasserstein GAN with Gradient Penalty (WGAN-GP) for facial image generation. The provided models are trained for 200 epochs, showcasing integration of techniques from key research papers. Deeper Networks and more Training can improve results.
GAN-Digit-Generator
This PyTorch-based GAN generates MNIST-like handwritten digits. It features a generator with transposed convolutions to create 28x28 images from noise and a convolutional discriminator to classify images as real or fake. The model trains adversarially using standard GAN loss to improve both generator and discriminator performance.
Hand-Sketch-Recognition--InceptionV3
The Hand Drawn Sketch Classification project uses PyTorch to classify hand-drawn sketches. It evaluates model Inception_v3, with Inception_v3 achieving the highest accuracy of 57%. The repository features scripts for dataset management, model training, and evaluation.
Library_Management_System_Java
A command-line Java application for managing library books and users. Features include adding, displaying, borrowing, returning, and searching for books and users. Data is stored in text files, and no external dependencies are required for running application.
Road_Lane-Detection-OpenCV
Real-time Lane Detection: A digital Image Processing project detecting lanes in live video feeds. Utilizes OpenCV & Python for accurate lane identification.
MuhammadShavaiz's Repositories
MuhammadShavaiz/Advanced-Data-Structues-FIFA_Dataset
This C++ project showcases efficient data management for FIFA World Cup data from 1930 to 2018. It uses AVL Trees for fast searching, Hash Tables for quick access, and Singly Linked Lists for easy traversal, optimizing both memory usage and retrieval speed.
MuhammadShavaiz/Aeon-Sentinels-Unity-Game
A Unity-based tower defense game using C#, ShaderLab, and HLSL. Features dynamic difficulty scaling and single-player mode, utilizing Unity’s physics and rendering systems.
MuhammadShavaiz/AI_learning
Google Colab notebooks showcasing PyTorch implementations and experiments. Covers deep learning techniques, including neural networks and NLP concepts.
MuhammadShavaiz/Clothes-Generator-GAN-s
The Clothes-Generator project uses TensorFlow and GANs to create synthetic clothing images from the Fashion MNIST dataset. Leveraging TensorFlow’s framework, the project explores advanced image synthesis techniques to improve generated image quality.
MuhammadShavaiz/DL-Visual-Question-Answering
The Visual Question Answering (VQA) project features a model with a simple GUI that handles both images and videos. It uses OpenAI's CLIP for encoding images and questions and GPT-2 for decoding embeddings to answer questions based on the VQA Version 2 dataset, which includes 265,016 images with multiple questions and answers.
MuhammadShavaiz/Face-Generation-Wasserstein-Gan-s-GP
This project demonstrates a GAN built with PyTorch, using a subset of 5000 CelebA images. It leverages Wasserstein GAN with Gradient Penalty (WGAN-GP) for facial image generation. The provided models are trained for 200 epochs, showcasing integration of techniques from key research papers. Deeper Networks and more Training can improve results.
MuhammadShavaiz/GAN-Digit-Generator
This PyTorch-based GAN generates MNIST-like handwritten digits. It features a generator with transposed convolutions to create 28x28 images from noise and a convolutional discriminator to classify images as real or fake. The model trains adversarially using standard GAN loss to improve both generator and discriminator performance.
MuhammadShavaiz/Hand-Sketch-Recognition--InceptionV3
The Hand Drawn Sketch Classification project uses PyTorch to classify hand-drawn sketches. It evaluates model Inception_v3, with Inception_v3 achieving the highest accuracy of 57%. The repository features scripts for dataset management, model training, and evaluation.
MuhammadShavaiz/Library_Management_System_Java
A command-line Java application for managing library books and users. Features include adding, displaying, borrowing, returning, and searching for books and users. Data is stored in text files, and no external dependencies are required for running application.
MuhammadShavaiz/Road_Lane-Detection-OpenCV
Real-time Lane Detection: A digital Image Processing project detecting lanes in live video feeds. Utilizes OpenCV & Python for accurate lane identification.
MuhammadShavaiz/LSTM-MUSIC-GENERATOR
In this project, I have created a LSTM model that predicts/generates music based on the initial notes of time series.
MuhammadShavaiz/MuhammadShavaiz
Config files for my GitHub profile.
MuhammadShavaiz/Portfolio
This repository hosts a portfolio website built with React, HTML, CSS, and JavaScript, and deployed via GitHub Pages. It highlights modern web development skills and practices, featuring clean code and effective deployment strategies.
MuhammadShavaiz/SearchEngineJava
MuhammadShavaiz/SEMANTIC-IMAGE-SEGMENTATION
This repository implements and evaluates SegNet, U-Net, and models from segmentation_models.pytorch for vehicle segmentation using PyTorch. It showcases these models' capabilities in accurately segmenting vehicles in complex scenes, demonstrating cutting-edge deep learning techniques in computer vision
MuhammadShavaiz/Shotes-React-Django-MiniProject
This repository features a Notes App built with React for the frontend and Django for the backend. It includes functionalities for creating, deleting, and automatically saving notes, with support for light and dark modes. It showcases a full-stack approach to modern web development.
MuhammadShavaiz/Spam_or_Ham_EmailDetection
The Spam Email Detection project uses Scikit-learn and Pandas to classify emails as spam or not, leveraging a dataset of 5,000 emails. It demonstrates practical statistical analysis and model training for effective email classification.