Machine Learning in Python with scikit-learn

This repository will contain the material for the Strata tutorial

Machine learning in Python with scikit-learn

by Andreas Mueller.

9:00am–12:30pm Tuesday, 09/27/2016

Data Science and Advanced Analytics

Location: 3D 12

It is recommended that you update the materials before the course, as they
might change in the days leading up to the conference.

Please bring a laptop with a working installation of Python (2.7, 3.4 or 3.5).
The following packages are required:

• scikit-learn >= 0.17
• matplotlib >= 1.3
• numpy >= 1.5
• IPython >= 4.0
• Jupyter Notebook >= 4.0

The easiest way to install all requirements is to install the free Anaconda Python distribution:
https://www.continuum.io/downloads (OS X, Windows, Linux)

Please download the material prior to arriving to the tutorial, and make sure
you can run the notebooks. To run a notebook, start Jupyter Notebook and
browse to the folder to which you downloaded it.

Outline

• Part 01 - Introduction to Scikit-learn
• Part 02 - Unsupervised Transformers
• Part 03 - Cross-validation
• Part 04 - Grid Searches for Hyper Parameters
• Part 05 - Preprocessing and Pipelines
• Part 06 - Working With Text Data
• Part 07 - Feature Union
• Part 08 - Out Of Core Learning
• Part 09 - Out Of Core Learning for Text