A typical ML workflow includes the following steps: data acquisition, data analysis, data preparation, data validation, model building, model training, model validation, training at scale, model inference, and monitoring.

The following figure shows a sample workflow:

Figure 2.              Machine learning workflow