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 modelsNeuralNetworks
: Julia module for implementing feed forward neural network models
The report is the file Project-2-Report.pdf