This article shows how to use a classification Machine Learning (ML) algorithm on the Iris Flower data set. It first creates a model based on a training data set and then scores the the test data set and gives the prediction of the flower based on the model.
- Install the following python libraries on the Incorta server using pip install if they are not installed.
pandas
numpy
matplotlib
seaborn
scikit-learn
- Import the schema zip from schema page and the data file from the UI . The Iris ML schema has two examples, one using pandas and other using incorta_ML .
- Load the IrisData table
- Now open the materialized views in a notebook and start running each of the paragraphs to see what it does.