neill-erasmus Goto Github PK

followers: 3.0 following: 27.0 repos: 27.0 gists: 0.0

Name: Neill Jean Erasmus

Type: User

Location: South Africa

Neill Jean Erasmus

💥 About Me

Driven and forward-thinking Computer Science student at Nelson Mandela University, with a strong passion for AI and machine learning. Equipped with a diverse skill set honed through academic pursuits and hands-on project experiences, as evidenced by my portfolio on GitHub.

I am deeply committed to leveraging technology to solve real-world challenges and drive positive change. Inspired by a dedication to innovation and excellence, I am constantly seeking to push the boundaries of what is possible in the field of AI and machine learning. With a keen understanding of the importance of user-centric design and a commitment to quality craftsmanship, I am well-positioned to contribute effectively to projects that demand both technical expertise and creative thinking.

As I embark on the next phase of my career journey, I am eager to apply my skills and passion to an organization that values innovation, collaboration, and a commitment to making a meaningful impact on the world through technology.

🚀 Skills

Programming Languages

Python
C#

Machine Learning

Regression
Classification
Clustering
Association Rule Learning
Reinforcement Learning
Natural Language Processing
Dimensionality Reduction

Deep Learning

Artificial Neural Networks
Convolutional Neural Networks
Recurrent Neural Networks
Self Organizing Maps
Boltzman Machines
Autoencoders

Artificial Intelligence

Q-Learning
Deep Q-Learning
Deep Convolutional Q-Learning
A3C (Asynchronous Advantage Actor-Critic)
Augmented Random Search

Computer Vision

Convolutional Neural Networks
Generative Adverserial Networks
Deep Convolutional Generative Adverserial Networks

🌱 Learning

Currently, I am really focused on improving my knowledge and practical ability in the areas listed in my skills section. I think it is important to keep learning because there's always more to know about everything. I believe that understanding things well is a constant journey, and I am working on not just being good but really understanding how everything works in the areas I am interested in.

Studies

I am currently pursuing my Bachelors Degree in Computer Science at Nelson Mandela University.

Udemy Courses

🔭 Projects

Check out some of my more notable projects:

📫 Contact

Let's connect! Feel free to reach out:

📈 GitHub Stats

🏆 GitHub Trophies

🤝 Let's Collaborate

I am interested on collaborating on any projects especially those that involve artificial intelligence!

Neill Jean Erasmus's Projects

2024-sproutopia

ad-selection

Using upper confidence bound reinforcement learning to determine the most effective advert in a dataset.

advertising-budgets-and-sales-regression

Created a multiple linear regression model to predict business sales returns, utilizing data from TV, radio, and newspaper advertising budgets. The model aids in optimizing advertising strategies for enhanced return on investment.

ai-generated-art

I created and trained a DCGAN (Deep Convolutional Generative Adversarial Network) to produce artificial portraits using a dataset containing more than 6000 images.

chrome-dino-automated

A python script to automate the chrome dino game.

credit-card-fraud

Using a self organizing map (SOM) to identify frauds in the credit card application process.

customer-personality-analysis

Leveraging K-Means clustering for insightful customer segmentation, enabling businesses to tailor products to specific customer types.

diabetes-classification

An artificial neural network-based model for diabetes prediction, leveraging machine learning techniques to analyze relevant health data and provide accurate predictions regarding the likelihood of diabetes.

file-cleanup-automation

A python script to automate the cleanup of junk files on your Windows system.

gender-classification

Comparing different classification models on the same dataset to classify people into genders based on several features.

keras

Deep Learning for humans

kung-fu

I utilized the A3C (Asynchronous Advantage Actor-Critic) algorithm to train a Deep Q-Learning (DQN) model, specifically tailored to solve the Kungfu gym environment.

lunar-lander

Trained a Deep Q-Learning agent to autonomously land a lunar module in OpenAI's Gymnasium Lunar Lander environment.

market-basket-optimization

Using the apriori association rule learning algorithm to identify goods commonly associated and purchased together.

neill-erasmus

Config files for my GitHub profile.

opencv

Open Source Computer Vision Library

pacman

I developed and trained a deep convolutional Q-learning model to enable an agent to successfully solve the Pacman gym environment.

pdf-to-audiobook

A python script that generates a short story and saves it in PDF format, before rendering an audio version and saving it as an MP3 file.

pytorch

Tensors and Dynamic neural networks in Python with strong GPU acceleration

restaurant-reviews

I trained a natural language processing (NLP) model to classify restaurant reviews as either positive or negative sentiment.

scikit-learn

scikit-learn: machine learning in Python

stock-price-prediction

This project employs LSTM, a type of recurrent neural network, to forecast stock market indices using historical data with features like date, open, high, low, close, and volume. By training the model on this dataset, it learns patterns to predict future market trends, aiding investors, traders, and analysts in decision-making.