The following diagram shows a typical workflow of an ML project:
Everything starts with the gathering of data. The data can come from various sources in various forms. Once the data is stored, we clean and preprocess it. Then we split our data into two parts. The first set is for training and the second set is for testing.
The training data, along with an ML algorithm, is used to build a data model. This data model is evaluated against the testing dataset that we have kept to one side. This results in a more accurate prediction. If the accuracy is acceptable, then we deploy this data model.
If new data comes in, we ...
