Jake Hawkesworth's repositories
tractor_case
Predicted tractor sales prices at auctions using multiple datasets. Applied linear regression, imputing missing data with k-Nearest Neighbors (kNN), and evaluated results using root mean squared logarithmic error (RMSLE score).
boomsetdataclean
Tool for cleaning event attendance data from Boomset
churn_case
Identified customers who are likely to churn using logistic regression and Random Forest classifiers. Evaluated results using Area Under the Curve (AUC) score.
kmeans
K Means algorithm coded in Scipy and Numpy, tested on Iris dataset.
kruegerreplication
A Replication of Alan Krueger’s "What Makes a Terrorist?"
MNIST
C-NN using Keras and Theano to categorize handwritten numbers, 99.3% accuracy
fancyimpute
Multivariate imputation and matrix completion algorithms implemented in Python
fraud_detection
A Gradient Boosted Model that triages fraudulent events, and recommends whether or not to flag a given event for further investigation.
MineSweeperNeuralNet
Teaching a neural net to play mine sweeper
models
Models and examples built with TensorFlow
NN_fun
Some simple examples showing off power of Neural Networks
scala_practice
Simple solutions to scala challenges
scikit-learn
scikit-learn: machine learning in Python
T-Watch
Real Time Twitter Sentiment Analysis Product
timetofailure
Kaggle competition and Linear Algebra practice