Aidan Stack's repositories
Detecting_Steel_Defects_with_Neural_Networks
This project explores using deep learning techniques to classify defects in images of steel in a real world business context. The entire project was done on AWS (S3 and Sagemaker). Designed and tuned artificial and convolutional neural networks. Achieved 85% accuracy in binary classification, and 79% accuracy in multiclass classification.
CodeWars_Puzzles
CodeWars.com is a website that gives short programming challenges. This is a collection of a few of my Solutions.
Cycling_Classification_Project
Using 1.8 million rows of traffic collision data from NYC's Open Data initiative, we ran thousands of classification models to discover which variables made collisions lethal for cyclists. The data was cleaned using Pandas, then Scikit-learn was used to instantiate and gridsearch Decision Tree, Random Forest, Logistic Regression, and K Nearest Neighbor models. Logistic Regression proved the most reliable, and was tuned for recall, as fatal collisions made up only .04% of all collisions involving cyclists. The variables most associated with lethal cycling collisions were then extracted back out as actionable insights.
Zillow_Real_Estate_Time_Series_analysis
Using Zillow data, we utilized Statsmodels' ARIMA functionality to forecast future home prices for 26 metro areas in NY State.
Leetcode.com_Problems
A few of my completely original solutions to Leetcode.com programming challenges
SQL_Handbook
My own notes on the basics of SQL