1991sig / FYS-STK4155-Project-2

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Project-2

Repo for project 2 in FYS-STK4155 - Classification on the Credit Card Dataset.

I have tried out two models:

  • Logistic Regression using Fisher Scoring
  • Neural Network using backpropagation

Both of the models have been implemented from scratch in Julia.

Data contains the data files that have been split into train, test, and split, as well as the means and standard deviations of the covariates. Credit Card Data.ipynb details the splitting and cleaning of the data.

  • LogisticRegression: Julia module for implementing logistic regression models
  • NeuralNetworks: Julia module for implementing feed forward neural network models

The report is the file Project-2-Report.pdf

About


Languages

Language:Jupyter Notebook 81.7%Language:TeX 8.4%Language:Julia 8.1%Language:R 1.8%