From the course: NLP with Python for Machine Learning Essential Training
Unlock the full course today
Join today to access over 22,600 courses taught by industry experts or purchase this course individually.
Box-Cox power transformation - Python Tutorial
From the course: NLP with Python for Machine Learning Essential Training
Box-Cox power transformation
- [Lecturer] So now that we've identified a feature to transform, it's time to tackle the elephant in the room. What is a transformation? A transformation is a process that alters each data point in a certain column in a systematic way that makes it cleaner for a model to use. For instance, that could mean squaring each value, or maybe taking the square root of each value in a given column. So let's say a distribution for a certain feature has a long right tail like this one does in this image. Then the transformation would aim to pull that tail in to make it a more compact distribution like we see in the example here using a log transformation. We do this so that the model doesn't get distracted trying to chase down outliers in that tail. The series of transformation that we'll be working with are call the Box-Cox Power Transformations. This is a very common type of transformation. The base form of this type of transformation is y to the x power, where y is the value in an individual…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.