Titanic: Learning From Disaster(Kaggle)

A python algorithm to solve Kaggle's 'Titanic: Learning From Disaster' machine learning problem using Radomized Forests. It gives a score of 0.79426 (Top 15%).

Install

All the code is contained within the 'titanic.py' scipt. It could be manually downloaded or cloned.

The script requires the sklearn and numpy packages which can be got using pip

pip install numpy

pip install scikit-learn

Usage

The script requires three parameters to run
-T - Your Training File
-t - Your Test File
-r - The final result file in proper format

An example of its usage within command line:

Python titanic.py -T /Users/YourUsername/Kaggle/Titanic/train.csv -t /Users/YourUsername/Kaggle/Titanic/test.csv -r /Users/YourUsername/Kaggle/Titanic/result.csv

References

Scikit-Learn Docs
Machine learning - Random forests