Machine learning process lifecycle and solution architecture

In this section, we will discuss the machine learning implementation process and solution architecture:

1. The first step toward defining the solution architecture is defining the problem statement, which includes defining the goal, process, and assumptions.
2. Determine what problem type is this problem classified under? Whether it is a classification, regression, or optimization problem?
3. Choose a metric that will be used to measure the accuracy of the model.
4. In order to ensure the model works well with the unseen data:
   1. Build the model using training data.
   2. Tweak the model using test data.
   3. Declare an accuracy based on the final version.

The following figure explains the flow and architecture of the underlying system: