Retraining and Saving an ML.NET Model

With ML.NET, we can train and retrain the model from the code. Whether we're training a new model or retraining an existing one, the process is largely the same. We would execute the same methods in the core library. Then, once the model is trained or retrained, we can save it in a file for consumption.

Retraining and saving a model in ML.NET

Sometimes it’s necessary to retrain the model. For example, the original dataset might have been too small or the data inside it was inaccurate. Perhaps we also want to clear up the data further because certain features don’t make the model any more accurate, but their presence significantly increases the training time.

In this lesson, we'll go through the process of training and retraining the model by using ML.NET. We'll do so with the help of the following playground: