This repository contains a series of Jupyter notebooks that demonstrate various applied AI concepts. The projects cover machine learning, deep learning, and building a neural network from scratch as part of my minor in applied AI.
We were instructed to cover a wide range of topics in AI, encompassing approximately 50 different areas. However, the implementation of these projects was left open-ended, allowing for creative freedom in their development, meaning that the projects below are my own ideas and implementations.
Note that when viewing the notebooks on GitHub, that the links which navigate inside notebooks (Table of Contents + others) do not work.
- Predicting Life Expectancy using Linear Regression - A linear regression model to predict life expectancy based on a dataset from the World Health Organization.
- Binary Classification of Organic-Non-Organic Images - A binary classification model to classify images of organic and non-organic waste.
- Songe Genre Multi-Class Classfication - Trying to classify a collection of 30000 spotify songs into 6 genres.
- Glass Identification with PCA - Exploring the UCI Glass Identification dataset using PCA.
- Clustering of Wine Data - Clustering the UCI Wine dataset using KMeans and Agglomerative clustering.
- Game Strategy using Reinforcement and Transfer Learning - Custom GridWorld implementation that uses reinforcement learning to train an agent to and transfer learning for scaling to a larger grid.
- Text Classfication using an RNN and BERT - Classifying tripadvisor hotel reviews using an RNN and a pre-trained BERT model.
- Anomaly Detection using Autoencoders and Adversarial Training - Detecting anomalies in the MNIST dataset using autoencoders and improving the model robustness using adversarial training.
- Fruit Classification using DNN, CNN, and ResNet50 - Classifying images of fruits using a Deep Neural Network, Convolutional Neural Network, and ResNet50.
- Building a Neural Network from Scratch - Building a neural network from scratch using only numpy and implementing forward and backpropagation.
To run the notebooks locally (though they can be viewed directly on GitHub), follow these steps:
- Clone the repository:
git clone https://github.com/yourusername/ml-dl-projects.git
cd ml-dl-projects- Create a virtual environment (Optional but recommended as per PEP 405):
python -m venv venv
source env/bin/activate # On Windows, use `.\env\Scripts\activate`- Install the required packages:
pip install -r requirements.txt- Start Jupyter:
jupyter notebook