Example 1Lesson 26Free preview

Ex1 Training

This lesson is part of A.I. · All levels

Summary

Training is conducted with model.fit(xs, ys, {epochs: n}) which repeats the learning process n times over the entire data. Each epoch, the weight is adjusted to reduce the loss. After the training is completed, model.predict() is used to estimate new values, e.g. y for x outside the training data.

Key points

Model.fit() receives data input, expectation output, and training options.
Epochs determine how many times all the data is passed during training.
The training results (loss per epoch) are stored in the promise/history.
Model.predict() is used to predict new data output.
The steady decline in weight indicates that the model is better at adjusting.

Full lesson

Code examples and quizzes

Full lessons, code examples, and quizzes are available to subscribers. 5 questions.