Intro to ML - Guide for Data Scientists [Recommendation]
Those of you following the machine learning video tutorials know that I often mention books that I follow and/or get inspiration from. It's not that every idea or thought that I put in these tutorials spring from my mind alone; I don't and I won't reinvent the wheel (hopefully).
Anyhow, a major source of inspiration for the tutorials has been the machine learning book by Andreas Muller and Sarah Guido, Introduction to Machine Learning with Python: A Guide for Data Scientists. And I still have quite a lot of material to cover from it (and from other books), but I have to figure out what to prioritize on.
Those of you with good programming skills and with knowledge of linear algebra, probabilities, and statistics should be able to jump right in and work through the book yourselves with relative ease. Which is why I kindly recommend it.
In short, here are a few of the many topics covered in the book:
- essential libraries and tools for ML
- supervised learning: classification and regression; what to do when overfitting and underfitting
- algorithms and models for supervised ML: KNN, linear models, trees and forests, SVMs, neural nets
- unsupervised ML: types, challenges, feature extraction and dimensionality reduction
- evaluating and improving your model
- chains and pipelines; these are very important when you're evaluating/looking for a good model out of many
- working with text data - exemplified approach.
If it helps, you could, at the same time, go through my video tutorials on machine learning with scikit-learn, starting with the first video tutorial [here]. There are currently 28 videos tutorials, so you might have some catching-up to do. :)
To stay in touch with me, follow @cristi
Cristi Vlad, Self-Experimenter and Author
Great book! I own a copy myself, and is really one of my main resources
Well...I hope I do. Or, I guess I hope I am at least an above average programmer. ;)
Thanks for the post.
Thank you, this looks awesome. :-)
My knowledge of python is rudimentary, to say the least, however, have a strong background in stats, would you suggest I start with pure Python first or try to jump straight in?
EdX also has a number of awesome free moocs on the subject matter, that may be of interest to some.
https://www.edx.org/course?search_query=machine+learning
I'd suggest pure python first, until you get a good grasp of the basic. look into sololearn.com as a platform to learn python for free, hands-on. and practice everyday.
Amazing, will definitely check it out. Thank you for the advice mate. :-)
Thanks for all the videos and all the articles on Machine Learning. I will go through the lessons one at a time soon
good luck! and let me know if you need help.
Machine learning is definitely the you're and I'd love to get in to it
I dont understand.
Well see its totally the you're. Makes sense now?
no :D
Why have i never heard about this before.
Thank you for this amazing article @cristi.
Just forward you are a true steemer :)
am I?
Yeap :)
how do you know?
My cousin has a friend who uses dark magic, he told me 😂
Many thanks for sharing! Upvoted and followed you!
Thanks so much @cristi for your valuable contribution to steemit community , I've been following and getting up to speed in ML space. I was going through your videos, is it possible to point me if there are any cluster analysis ( behavioral patterns ) or similar that you've done either in R or scikit ( preferably R ) . I'm looking at clustering right now and how to approach with the data we have , would be great if you can help me these.