Convolutional Neural Networks in Python: Master Data Science and Machine Learning with Modern Deep Learning in Python, Theano, and TensorFlow (Machine Learning in Python) by LazyProgrammer

Author:LazyProgrammer [Unknown]
Language: eng
Format: epub, pdf
Published: 2016-05-15T05:00:00+00:00

So in Theano, our first filter has the dimensions “num_feature_maps”, which you can think of as the number of kernels or filters we are going to create, then it has “num_color_channels”, which is 3 for a color image, and then the filter width and height. I’ve chosen to use 5 since that’s what I usually see in existing code, but of course this is a hyperparameter that you can optimize.

# (num_feature_maps, num_color_channels, filter_width, filter_height)

W1_shape = (20, 3, 5, 5)

W1 = np.random.randn(W1_shape)

b1_init = np.zeros(W1_shape[0])

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.