A pipeline
is a series of steps that transform data and apply a machine learning algorithm to make predictions. Understanding the different steps of a pipeline is crucial for successfully building and deploying machine learning models. In this thread, we will discuss some ways to examine the steps of a pipeline using Sci-kit learn
.
Loading "dataset" and creating "pipeline":
First, let's load the iris dataset and split it into training and testing sets. Then, create a pipeline that applies linear regression to the data and fit the pipeline to the training data. Let's understand below:To examine the steps of the pipeline, we can use the following methods:
1. Using "pipeline.steps" :
It returns a list of tuples, where each tuple contains the name of the step and the actual estimator object. For example:2. Using "pipeline.named_steps" :
It returns a dictionary that maps step names to estimator objects. For example:3. Using "pipeline.get_params" :
It returns a dictionary of parameter names mapped to their values for each step in the pipeline. For example:4. Using "pipeline.score" :
It method returns the R^2
coefficient of determination for the pipeline. For example: