Decision tree is a decision making tool in python for data science that uses a flowchart or is a model of decisions and all the possible outcomes, including cost and utility.
The decision tree algorithm comes under a supervised learning algorithm. It works efficiently for both continuous and categorical output variables.
The results are reflected on the branches of the node, and the nodes will have either:
- Conditions or decision nodes
- Result or end nodes.
The branches represent truth or falsity of a statement, and it makes a decision which is based on it. The below example represents a decision tree that will evaluate the smallest of three numbers.
Decision tree regression:
It observes the features of an object and then trains a model in the structure of a tree for predicting data in the future for producing meaningful continuous output. By continuous output we mean the result is not discrete.
Discrete output examples: We will make a model for the prediction of rain on a particular day.
Continuous output example: It is a profit prediction model which will state the probable profit that can be generated from the sale of a product.
In this, the continuous values are predicted with a decision tree regression model.
Now we will observe the Step by step implementation.
- Step 1: Importing the required libraries.
- Step 2: Initializing and printing the data set.
- Step3: Selecting all the rows and column 1 from the data set to “X”.
- Step 4: Selecting all rows and column 2 from the data set of “y”.
- Step5: Fitting the decision tree regressor to the data set.
- Step 6: Prediction of a new value.
- Step 7: Visualization of the result.
- Step8: The tree is exported and shown n a tree structure below.