24 October, 2016 by admin
Slides for “A practical introduction to machine learning with sklearn”
I have published the slides for my codelab A practical introduction to machine learning with sklearn:
The Python code associated to the codelab is available as a single zip file here. The code has been tested with Python 3.5 and Python 2.7. If you use the latter, you may encounter a SciPy bug when loading the MAT file for the first example. In this case, please downgrade your SciPy library to version 0.16.0. For example, if you are using Anaconda you can run:
conda install scipy==0.16.0
The three exercises are as follows:
- Music classification using a customized version of the Garageband dataset.
- Image segmentation using a clustering algorithm.
- Joke recommendation using a nonnegative matrix factorization algorithm. For this exercise, you should download the dataset 1_1 of the Jester collection. Note that this solution is not optimal, since missing ratings are considered as being identical to low ratings in the solver. It is only given as informal introduction to the topic.