Skip to content

A from scratch implementation of a logistic regression Machine Learning model using regularization

Notifications You must be signed in to change notification settings

henriquearaujo-98/logistic-regression-implementation

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

4 Commits
 
 
 
 

Repository files navigation

Regression Model Development

This project contains a Python implementation of a logistic regression model developed from scratch, using mathematical formulas and optimized with a custom implementation of the gradient descent algorithm.

The model was developed for the Breast Cancer Dataset from scikit-learn (sklearn).

For the sake of data visualization, only the features "Mean Radius" and "Mean Texture" of the dataset were used. Further more, only 1/6 of the dataset was used to induce a model's overfitting. This was done so that there could be an accuracy comparison between the regularized model and the unregularized model.

No Machine Learning Framework was used in the making of this project as Scikit-learn was only used to load the dataset into NumPy arrays.

Requirements

To run the project, you will need the following Python packages:

  • NumPy
  • Pandas
  • Matplotlib
  • Scikit-learn (only to load the Boston Housing Dataset)

You can install these packages using pip, conda, or any other package manager.

Files

The project contains the following file:

  • solution.ipynb: a Jupyter Notebook with the code and documentation for the project.

About

A from scratch implementation of a logistic regression Machine Learning model using regularization

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published