Chevron Left
Back to Machine Learning Feature Selection in Python

Learner Reviews & Feedback for Machine Learning Feature Selection in Python by Coursera Project Network

4.0
stars
72 ratings
12 reviews

About the Course

In this 1-hour long project-based course, you will learn basic principles of feature selection and extraction, and how this can be implemented in Python. Together, we will explore basic Python implementations of Pearson correlation filtering, Select-K-Best knn-based filtering, backward sequential filtering, recursive feature elimination (RFE), estimating feature importance using bagged decision trees, lasso regularization, and reducing dimensionality using Principal Component Analysis (PCA). We will focus on the simplest implementation, usually using Scikit-Learn functions. All of this will be done on Ubuntu Linux, but can be accomplished using any Python I.D.E. on any operating system. We will be using the IDLE development environment to demonstrate several feature selection techniques using the publicly available Pima Diabetes dataset. I would encourage learners to experiment using these techniques not only for feature selection, but hyperparameter tuning as well. Note: This course works best for learners who are based in the North America region. We’re currently working on providing the same experience in other regions....
Filter by:

1 - 12 of 12 Reviews for Machine Learning Feature Selection in Python

By Christian P

Aug 2, 2020

Course was useless. Videos of instructor writing out some lines of code, and mispronouncing words he must be reading from a script. The point it is to understand how to interpret and use the results of the code. He simply glances over it with "here's your output!". One example even failed when he used incorrect syntax, and he closed it, and moved onto next example. Another example had him actually saying "well this isn't a good example for this", but then proceeds to teach with this "bad example". Clearly too lazy to put together a good example for a course.

Worst class taken on Coursera - can't believe I paid for it...

By Jarin t t

Jul 20, 2020

There was very little explanation. There was nothing to do for me. How can it be a project centered course! I wonder.

By Yaron K

Apr 15, 2021

Code snippets showing a number of feature selection and dimensionality reduction techniques that are implemented in Python procedures. Their weren't instructions on how to get the code to run in the Rhyme desktop. However for this course it wasn't that important - I simply wrote down the names of the modules for future use.

Pro : Short and to the point. Within an hour you'll have seen a number of Python feature selection and dimensionality reduction procedures with a short demo for each one.

Con: If you want to use a technique well - you'll need to look it up and see an in depth explanation of the parameters and results. Looked up SelectKBest, RFE, etc. and indeed found excellent articles. BUT you need to first know that they exist and why they exist - and this is why I gave the course 5 stars.

By GHEORGHIȚĂ A

Jul 12, 2020

I learning something interesting, we can do many think with python, machine learning, AI

By Zaid R

Aug 5, 2020

Thanks, I truly appreciate all information

By sanjeev s

Aug 5, 2020

good

By Prateek S

Apr 6, 2021

The algorithms explained were at superficial level, and not in depth. Also, the codes were not very detailed, with no focus on variable data type. Hope the feedback is useful!

By Ali T

Aug 6, 2020

You get the basics on all the feature engineering algorithms you can use but no explanation on what the results mean or how to interpret them. I could have searched all of this up and it would have given me better explanations. Even worst the rhyme.com constantly kept on changing docks for me and I couldn't figure out why either. Definitely not worth the money.

By Thanh N N

Sep 2, 2020

very disappointed! I couldn't find the code for later videos to download, only the first video has downloadable script. There are errors in the script, the dataset Pima has less rows than the original data on UIC. What's the point of paying $9.99 for the bugged code? Each video is less than 5', only 8 videos in this project, not even any slides/ppt to describe/summarize methods... Why I have to pay for a bugged script and get nothing out of this?

By VR

May 26, 2021

C​oursera must check and verify the quality of the projects at first as the learners need to pay for each of them. Despite the project describption ... I could not find any useful points to recommend this project.

By Wolfgang P

Oct 1, 2020

Close to no explanation why are things done this way and not different. Certificate can hardly be used as signed by Prof. John Doe

By CHAN E C

Dec 20, 2020

Don't waste your time here.