Deep Learning Illustrated: A Visual, Interactive Guide to Artificial Intelligence (Sal Lopez's Library) by Jon Krohn Grant Beyleveld and Aglaé Bassens

Author:Jon Krohn, Grant Beyleveld and Aglaé Bassens
Language: eng
Format: epub
Publisher: Addison-Wesley Professional
Published: 2020-06-15T00:00:00+00:00

Also thanks to cheap compute, we are subsampling activations only once (with a max-pooling layer), whereas LeNet-5 did twice.16

We leverage innovations like ReLU activations and dropout, which had not yet been invented at the time of LeNet-5.

16. There is a general trend in deep learning to use pooling layers less frequently, presumably due to increasingly inexpensive computation costs.

If you’d like to follow along interactively, please make your way to our LeNet in Keras Jupyter notebook. As shown in Example 10.1, relative to our previous notebook (Deep Net in Keras, covered in Chapter 9), we have three additional dependencies.

Example 10.1 Dependencies for LeNet in Keras

Click here to view code image

import keras from keras.datasets import mnist from keras.models import Sequential from keras.layers import Dense, Dropout from keras.layers import Conv2D, MaxPooling2D # new! from keras.layers import Flatten # new!

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.