sujitpal / statlearning-notebooks

Python notebooks for exercises covered in Stanford statlearning class (where exercises were in R).

##StatLearning Notebooks##

IPython notebooks that implement the R code for the StatLearning: Statistical Learning online course from Stanford University taught by Profs Trevor Hastie and Rob Tibshirani. The original code for the classes were written in R.

List of Notebooks:

• Chapter 2: Basic Operations
• Chapter 3: Linear Regression
• Chapter 4: Classification
• Chapter 5: Cross-Validation and Bootstrap
• Chapter 6: Feature Selection
• Chapter 7: Nonlinear Models
• Chapter 8: Decision Trees
• Chapter 9: Support Vector Machines
• Chapter 10: Unsupervised Methods

The notebooks are also accessible from A gallery of interesting IPython Notebooks under the Statistics, Machine Learning and Data Science section. More information in my blog post.

This project was linked from Statistics.md of this project containing many curated lists of links to Python scientific programming resources.

statlearning-notebooks's Issues

LOOCV calculation Error

In loo_shortcut(X, y) function, acorrding to the ITSL,
mse = ((y[row] - ypred[row]) / (1 - hi))**2 should be right

License?

Thanks for making a repository for this! I would like to incorporate some of this code for educational purposes. Just to be safe, could you choose a license? MITLicense would be appreciated!

Calculate hi value in Chapter 5 and plot margin in chapter9

First, wonderful work for people to learn how to use python in statistics.
I think it should be power 2 and should not be hi = ((1 / nrows) + (np.sum(X[row] - xbar), 2) / xsum)[0] .

In chapter 9, to plot the Maximum margin separating hyperplane, I use code from here In your code, I think the c should equal to -reg.intercept_ / reg.coef_[0][1] but I cannot make it right by using SVR. These two methods may define reg.coef_ in different ways.