abdelrhmanghreeb Goto Github PK

aoc_id

Arabic Dialect Identification on AOC data.

bert

TensorFlow code and pre-trained models for BERT

bilm-tf

Tensorflow implementation of contextualized word representations from bi-directional language models

cervix-cancer-detection-kaggle-competition

This is my code to contribute in kaggle cervix cancer detection competition. I used pretrained xception model

doccano

Open source text annotation tool for machine learning practitioner.

elasticsearchlocalhosting

fully-connected-neual-network---native-implementation

This is an implementation of fully connected neural network on CIFAR10 data using numpy only. This is done as a proof of concept.

generative-ai-with-llms

In Generative AI with Large Language Models (LLMs), you’ll learn the fundamentals of how generative AI works, and how to deploy it in real-world applications.

movie-reviews-sentiment-analysis

This is a Sentiment Analysis project on movie reviews from IMDB dataset, which is a typical NLP problem. My A neural network and pre-trained word vectors from Glove gave me scores compatible to state of the art results very easily. This is an experimental project that is very much like real world problems. Wish you enjoyable reading.

movies-recommender-system

This is a recommender system applied in two different approaches: 1) Simple model 2) Neural Network based model Both models outperform the academic benchmarks with few epochs of training, with the second gives state of the art like results. I use the famous Movielens mini dataset in this project, which has users' ratings of movies. The aim of the project is to showcase a simple recommender system which is able to predict which movies certain user would probably like most by looking at what similar users liked. I used keras library as the core backbone for this project.

pima-dataset-classification-using-pca

The goal of the project is to use principal component analysis to perform a classification task for which the training data is fewer than what is needed for a deep learning model. I worked on the Pima Indians Diabetes Data Set (https://www.kaggle.com/dssariya/pima-indians-diabetes-data-set) I successfully implemented the model and a comparison with a solution without PCA gave useful insights in understanding the results.

privategithubcopilot

A self-hosted github copilot guide using oobabooga webui

reviews_tone_api

A fully working Flask API that uses IBM Watson's tone analyzer to index hotel review tones to a Elastic Search.

rtd_tutorial

spot-distracted-drivers---state-farm-kaggle-competition

State farm posted a competition on Kaggle to spot different 10 types of unsafe driving attitudes. In this project, I try to reach competitive results to those on the leader board of the competition. Using convolutional neural network based model, alongside with data augmentation, dropout, and pseudo labeling, we were able to reach very competitive results in this typical image classification problem. Because I have had no access to capable gpu at this time, I experimented on just small sample of the whole data and appended Jermey Howard's results to my notebook to give a sense on results that can be easily reached.

temp_repo

twiquery

Tweets scraper enables you to do advanced search for tweets like advanced search twitter web page, save the collected tweets.