From the course: Learning TensorFlow with JavaScript

Unlock the full course today

Join today to access over 22,600 courses taught by industry experts or purchase this course individually.

Layers introduction

Layers introduction

From the course: Learning TensorFlow with JavaScript

Start my 1-month free trial

Layers introduction

“

- [Instructor] When you start doing deep learning, the concept of layers becomes very important. And in TensorFlow.js, you can build your model out of layers to construct your type of neural network. Layers are the building blocks and basically performs operations with its inputs and then output the data to be ready for another layer or final output. Let's explore the API docs to see which ones are available and then build a simple sequential layer. If you go back to the website, to js.tensorflow.org, and you are in the API references, once you get there, scroll until you see LAYERS. It should be roughly after the models and you'll see all the layers that exist. You have the basic layers where you can do activation, dense, dropout, embedding. You have the events activation when you can do an elu. You also have the convolutional layers, the merge, normalization, pooling, and so on, so forth. Basically the layers pulls from the output of the previous layer and then does some computation…

Contents