Hands-On Deep Learning with TensorFlow by Dan Van Boxel

Author:Dan Van Boxel [Boxel, Dan Van]
Language: eng
Format: azw3
Publisher: Packt Publishing
Published: 2017-07-31T04:00:00+00:00

So we know where to start let's look at the example image using the following code:

# Let's take a look import matplotlib.pyplot as plt plt.ion() # Original plt.matshow(image[0]) plt.colorbar()

The 0 in the preceding code is just because of the strange shaping, there are not really multiple data points. You can see that the values on the diagonal are larger than the other values, just to be distinct from being purely random:

Let's take a look at the first output feature, recall that the output H has a shape of 1,10,10,2, because there's 1 data point, 10 pixels in width and height, and 2 features. So to grab the first, we want all the pixels and the zero with filter. Okay, that's interesting.

# Conv channel 1 plt.matshow(H[0,:,:,0]) plt.colorbar()

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.