Machine learning techniques are finding widespread application in industry for data-driven analysis and decision making. Software product vendors increasingly need to support data analysis use cases in their products. However integrated software engineering of ML components with software products or systems has not received much attention. This paper presents a feature-model based method and framework for integrating ML development into software product lines from the requirements stage. Our approach includes auto-generation of ML code from specifications to make ML development more accessible for product engineering teams and also enable configurability of ML solutions through their lifetime. We illustrate the approach using a real-world customer ticket classification case study.