Advanced Machine Learning with scikit-learn
Tools and Techniques for Predictive Analytics in Python
By Andreas C. Müller
Publisher: O'Reilly Media
Final Release Date: September 2015
Run time: 3 hours 43 minutes

In this Advanced Machine Learning with scikit-learn training course, expert author Andreas Mueller will teach you how to choose and evaluate machine learning models. This course is designed for users that already have experience with Python.

You will start by learning about model complexity, overfitting and underfitting. From there, Andreas will teach you about pipelines, advanced metrics and imbalanced classes, and model selection for unsupervised learning. This video tutorial also covers dealing with categorical variables, dictionaries, and incomplete data, and how to handle text data. Finally, you will learn about out of core learning, including the sci-learn interface for out of core learning and kernel approximations for large-scale non-linear classification.

Once you have completed this computer based training course, you will have learned everything you need to know to be able to choose and evaluate machine learning models. Working files are included, allowing you to follow along with the author throughout the lessons.

Table of Contents
Product Details
Recommended for You
Customer Reviews

REVIEW SNAPSHOT®

by PowerReviews
oreillyAdvanced Machine Learning with scikit-learn
 
4.0

(based on 1 review)

Ratings Distribution

  • 5 Stars

     

    (0)

  • 4 Stars

     

    (1)

  • 3 Stars

     

    (0)

  • 2 Stars

     

    (0)

  • 1 Stars

     

    (0)

Reviewed by 1 customer

Displaying review 1

Back to top

(2 of 2 customers found this review helpful)

 
4.0

Great resource for a structured approach

By Divya

from New York, NY

Verified Reviewer

Comments about oreilly Advanced Machine Learning with scikit-learn:

Note: I am a beginner in Machine Learning and have primarily used scikit-learn for supervised learning tasks. I received access to this video after winning a raffle at a technical meetup.

I found all videos to be quite useful as a reference for various facets of working with a learning task - starting from preprocessing of data, to choosing and comparing different algorithms, as well as reporting the performance. The videos also cover how to deal with categorical and text data. For the algorithms, there are detailed sections on SVM and Random forest. The choice of topics covered in these videos is quite apt as I found the topics of data preprocessing and working with categorical data to be highly relevant for the initial parts of a project. Cross-validation and the section on metrics were useful for the evaluation. The section on pipelining is quite detailed and useful when trying different approaches. I was not as familiar with the efficient methods introduced in the section on EstimatorCV Objects and found it to be insightful. I have not used scikit-learn for regression-based tasks, yet I found the information provided in the introductory lectures as well as the few basics covered while the methods are being discussed, to be easy to follow.

The videos are well paced and instructive. The author does a great job of outlining the general approach for the interface of the models as well as the estimators, which makes it easy to compare the approaches as well as adapt the approach for new models not covered in the videos. The use of Jupyter notebooks helps the tasks due to the easy interface. The author also provides caveats for when some approaches may not generalize or for using the tools in the correct manner.

Some aspects that I would like to see addressed in a future version is more support for pandas data frames and scikit-learn, as well as dealing with complex pipelines when processing data using pandas dataframes. As expected, the videos are not meant to be comprehensive for all the tasks, however the videos provide a great starting point for the different sub tasks one might need for supervised or unsupervised learning.

These videos provide a general overview of how to approach the sub tasks (for instance, overview of using pipelines, grid-search, data pre-processing). Overall, I would recommend the videos to a user looking for a structured approach (and pointers) to different aspects where scikit-learn can assist with learning-based tasks.

Displaying review 1

Back to top

 
Buy 2 Get 1 Free Free Shipping Guarantee
Buying Options
Immediate Access - Go Digital what's this?
Video:  $79.99
(Streaming, Downloadable)