Training a simple sklearn model within notebooks

The goal of this section is to create a Python script that will produce a simple model on top of the diabetes dataset that you registered in Working with datasets in Chapter 7, The AzureML Python SDK. The model will be getting numeric inputs and will be predicting a numeric output. To create this model, you will need to prepare the data, train the model, evaluate how the trained model performs, and then store it so that you will be able to reuse it in the future, as seen in Figure 8.1:

Figure 8.1 – Process to produce the diabetes-predicting model

Let's start by understanding the dataset you will be working with. The diabetes dataset consists of data from 442 diabetes patients. Each row represents one patient. Each row consists of 10 features (0 to 9 in Figure 8.2) such as age, blood pressure, and blood sugar level. These features have been transformed (mean-centered and scaled), a process similar...