A free course about implementing the most popular machine learning algorithms using only pure Python and Numpy.
First of all why would you want to learn to implement those algorithms than let a library like Pytorch handle them for you?
While a library hides the implementation details, if you’re really looking to understand how things work you have to go behind the covers. That is really especially useful if you are into Data Science.
As such the algorithms that are going to be implemented in the course are:
- KNN
- Linear Regression
- Logistics Regression
- Naive Bayes
- Perceptron
SVM - DecisionTree
- Random Forest
- Principal Component Analysis (PCA)
- K-Means
- AdaBoost
- Linear Discriminant Analysis (LDA)
The course is made by Python Engineer Patrick Loeber who constantly releases high quality courses and tutorials
on Python and ML, and you can expect the same sort of quality on this on too.
It is offered as a multi-part Youtube playlist or as a single piece full course. In any case the showcase of the 12 Algorithms it is composed of spans up to 5 hours in length.
The accompanying code can be found on the project’s Github repo. The project has the following dependencies:
- numpy for the maths implementation and writing the algorithms
- Scikit-learn for the data generation and testing.
- Matplotlib for the plotting.
- Pandas for loading data.
Note, however, that only numpy is used for the implementation. The others help in the testing of code and making it easy – instead of having to write that from scratch too. To follow along you just need basic Python, object-oriented programming and the basics of NumPy.
All in all, this is a very useful and excellent course on the fundamentals building blocks of Machine Learning. Totally recommended.
More Information
ML algorithms from Scratch on Github
Youtube playlists
Single piece full course
Related Articles
Triple Treat Machine Learning
The Year of AI Breakthroughs 2022
Take Google’s Machine Learning Crash Course
To be informed about new articles on I Programmer, sign up for our weekly newsletter, subscribe to the RSS feed and follow us on Twitter, Facebook or Linkedin.
Comments
or email your comment to: [email protected]
