Neural Arabic text diacritization

The random forest, FFNN, CNN and RNN models are developed to predict the movement of future trading price of Netflix (NFLX) stock using transaction data from the Limit Order Book (LOB).

Bimodal and Unimodal Sentiment Analysis of Internet Memes (Image+Text)

Deep Learning for file type Identification in Digital Forensics. We use the first, body, and last blocks of bytes on the disk to account for all possible scenarios and train the FFNN, CNN, GRU, and LSTM models. Afterward, we make predictions and evaluate the performance of each model.

Feed Forward Neural Network for Sentiment Classification and Language Modeling

Early mouse gesture recognition experiments / Delphi

Implementations of various deep learning pipelines

Simple FFNN using C++

Face Recognition by Eigenface method with the trained Feed Forward Neural Network and other classifiers applied to biometric attendance system functional on static-images.

Classification of acronyms and their long forms using an RNN (LSTM), CNN, and FFNN model. The experiments focused on the RNN and used different vectorisation methods and hyperparameters. Models were built with Keras and the notebook code runs on Google Colab.

A feed forward neural network (FFNN) is built to recognize the gray-scale images of hand-drawn digits from zero through nine using tensorflow.

Feedforward Neural Network (FFNN) from Scratch

A repository of assignments performed during the Advanced Machine Learning course.

Revolutionize text summarization with this Transformer model, leveraging state-of-the-art techniques. Trained on news articles, it produces concise summaries effortlessly. Explore cutting-edge capabilities for your summarization needs.

GlucoTrack is a machine learning and deep learning project focused on predicting a person’s risk level of diabetes

A simple and fast feed forward neural network library.

Web UI for the data behind PicPic, an automatic image selection tool for news articles

Fully Connected Forward Feed Neural Network

Feed Forward Neural Network (Regression)

End-to-end Deep Learning pipeline for network flow classification using the CICIDS2017 dataset. Implements Feed-Forward Neural Networks (FFNNs), handling class imbalance with weighted loss functions and analyzing feature bias. Includes extensive EDA, preprocessing, and regularization experiments to optimize model performance.