Machine Learning with scikit-learn with Data School
This is a free course offered by Data School.
The course is good for people who have completed Andrew Ng’s class on Machine Learning. You would start to realise how scikit-learn automates a lot of the low-level Octave code you implemented in Andrew Ng’s class from the simple gradient descent algorithm, to regularization, and the mind-boggling neural networks. Although I will be writing more on Neural Networks using Tensorflow (formerly known as Skflow).
Also, the tutorials offered by Data School are good for people who are new to scikit-learn and Python in general. If you do not want to watch his free open-source videos, you can check out my notes on all his tutorials.
- Introduction to Machine Learning
- IPython Introduction
- Exploring Iris Dataset
- KNN Classification Model
- Linear Regression Model
- Cross-validation for Parameter Tuning, Model Selection and Feature Selection
- Efficiently Searching Optimal Tuning Parameters
- Evaluating a Classification Model
I hope you found my notes useful! They are some variations on Github by others so feel free to use them too if you feel their explanations are clearer.
I will be moving on Pandas Series by Data School. Pandas plays an important role in data wrangling. Check out my review on this series soon!