This is the code developed to complete Georgia Tech CS 7641 (Machine Learning) assignment 1 (Supervised Learning).
The objective was to fit two datasets:
- A subset of Google's Quick, Draw! dataset here
- Credit card fraud detection using a dataset from Kaggle here
Using five classification algorithms:
- Decision Trees
- Support Vector Machines
- Artificial Neural Networks
- Boosting (using AdaBoost)
- K-Nearest Neighbors
The code is broken up into a Jupyter Notebook for each data-algorithm combination, each of which contains a few notes describing results. Supporting helper functions are contained in utilities.py, and data in the data folder (note that the data is compressed and must be uncompressed to run code).