Week 47: From Decision Trees to Ensemble Methods, Random Forests and Boosting Methods and Summary of Course

 

 

 

Preparing Your Data

  1. Shuffle your data
  2. Mean center your data
  3. Normalize the variance
  4. Whitening
  5. When to do train/test split?

Whitening is a decorrelation transformation that transforms a set of random variables into a set of new random variables with identity covariance (uncorrelated with unit variances).