Kaggle has a very nice walk-through of the Titanic data. The first two links are more towards processing the data, and the last uses scikit-learn's Random Forest.

https://www.kaggle.com/c/titanic/details/getting-started-with-python https://www.kaggle.com/c/titanic/details/getting-started-with-python-ii https://www.kaggle.com/c/titanic/details/getting-started-with-random-forests

In addition, Kaggle has a number of other "learning" competitions, where people post scripts, many of which utilize scikit-learn.

For example, this script analyzes San Francisco Crime data using Naive Bayes:

https://www.kaggle.com/sonuk7/sf-crime/prediction-with-bernoulinb

There are hundreds of other scripts that you can fork and modify to try different approaches.

One advantage of walking through scripts on Kaggle is there is a very active forum, so you can ask specific questions about code.