Boilerplate example using IPython Notebook to solve simplest (sex-field only) Titanic challenge for Kaggle (this will get you started wtih the Kaggle competition)
This is the Titanic challenge: http://www.kaggle.com/c/titanic-gettingStarted/
For a longer and far more in-depth analysis using Pandas see: http://nbviewer.ipython.org/urls/raw.github.com/agconti/kaggle-titanic/master/Titanic.ipynb
You must download the data from Kaggle to run this Notebook.
This Notebook replicates this example but using Python and an IPython Notebook: http://www.kaggle.com/c/titanic-gettingStarted/details/getting-started-with-excel
Run the Notebook using:
$ ipython notebook --pylab=inline # start the Notebook
Go to the webbrowser, open the "kaggle_titanic_sexcolumnonly_decision_tree" example. Work through it, generate the csv output file at the end and upload to Kaggle. You can also view the generated decision tree in-line. The decision tree that you visualise will look like:
You will probably need the dependencies too:
$ pip install -r requirements.txt # uses pip to get all the Python bits
Please note that I can't help you setup your machine, you'll have to figure that out yourself. Feel free to send me a Pull Request if you have improvements to the documentation. This tutorial cannot say any more about how to solve the Kaggle competition as that would be against the rules.