January 2018
Intermediate to advanced
310 pages
7h 48m
English
A perceptron is a single-layer neural network. The concepts covered in this chapter, such as a fully connected layer, activation function, Stochastic Gradient Descent, logits, one hot encoding, softmax, and cross-entropy will be useful here. You will learn how to define these components of neural networks in TensorFlow and use the network to train the MNIST data.
Read now
Unlock full access