Training an ML Model

Model overview

• Dataset: We will use the classic iris.csv dataset as input. The dataset has six columns with various flower features. The model classifies them into one flower species.

• Training model: We will use SVMSupport Vector Machine for classification in this lesson.

We need two critical components for running this job:

• job.yml: This YAML file contains the parameters and input required for running the job.
• main.py: This is the actual Python code to run.

The Iris dataset is almost like a “hello world” program for ML. Let’s look at the data definition. The file contains sepal_length, sepal_width, petal_length, and petal_width as the feature columns. The output column is a categorical attribute of the species name (Iris-versicolor/Iris-virginica/Iris-setosa).