In order to demonstrate the end-to-end process of building a predictive model (machine learning or supervised learning), we have created an easy-to-comprehend workflow. The first step is to design the problem, then source and prepare the data, which leads to coding the model for training and evaluation, and, finally, deploying the model. In the scope of this chapter, we will keep the model explanation to a bare minimum, as it will be covered again in detail in chapters 4 and 5.
The following figure describes the workflow required to build a predictive model starting from preparing the data to deploying the model:
Figure 3.5: Machine learning workflow.
Once we identify the domain of work, brainstorming on the designing of the problem is carried out. The idea is to first define the problem as a regression or classification problem. Once that is done, we choose the right target variable, along with identifying the features. The target variable is important...