Skip to content

Machine learning model for predicting passenger survival on the RMS Titanic

License

Notifications You must be signed in to change notification settings

OliverSieweke/titanic

Repository files navigation

License: MIT Version Python 3.7 Code style: black Dependabot Documentation Status Binder

Iceberg Logo

Titanic

Welcome to Titanic! This project proposes a machine learning model to predict passenger survival on the RMS Titanic and was submitted to the Kaggle competition Titanic: Machine Learning from Disaster.

Data

The data was downloaded from Kaggle on the 26.04.20 and includes:

  • data/original/train.csv which contain the details of a subset of 891 passengers on board, including their survival status.
  • data/original/test.csv which contains the data for 418 additional passangers for which the survival is unknown.

User Guide

Viewing the Project

The various notebooks used in this project for exploratory data anlysis, visualizations and predictive modeling can be viewed on MyBinder (this may take some time in case no container is currently deployed):

Binder

Running the Project

If you have Python 3 installed, you may run the project on your local machine by executing the following commands from your terminal:

$ git clone https://github.com/OliverSieweke/titanic.git
$ cd titanic
$ pip install -r requirements.txt
$ jupyter notebook notebooks

Documentation

Documentation for the project can be viewed on Read the Docs:

Documentation Status

Developer Guide

Contributions are welcome! Please fork the project, make sure you have Python 3.7 installed and set up your local repository as follows:

$ git clone https://github.com/<path_to_your_fork>
$ cd <your_local_repository>
$ python3.7 -m pip install -r requirements-dev.txt
$ pre-commit install

This will install required dependencies and set up git hooks to ensure that your commits conform to the project's standards and code style.

About

Machine learning model for predicting passenger survival on the RMS Titanic

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published