Example 1Lesson 25Free preview

Ex1 Model

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

Summary

The model is made with tf.sequential() and then a dense layer is added as needed. For simple regression, just one layer with one neuron, configured with the loss meanSquaredError and the sgd optimizer. Once compiled, the model is ready to be trained with the data that has been prepared.

Key points

Tf.sequential() compose layers of neural networks in sequence.
It's suitable for single feature regression.
'meanSquaredError' is used for the number regression problem.
The 'sgd' optimizer performs a stochastic descent gradient to update the weight.
Model.summary() displays the structure and number of model parameters.

Full lesson

Code examples and quizzes

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