MCPcopy Index your code
hub / github.com/lisa-lab/DeepLearningTutorials / get_reconstructed_input

Method get_reconstructed_input

code/cA.py:189–194  ·  view source on GitHub ↗

Computes the reconstructed input given the values of the hidden layer

(self, hidden)

Source from the content-addressed store, hash-verified

187 W, (1, self.n_visible, self.n_hidden))
188
189 def get_reconstructed_input(self, hidden):
190 """Computes the reconstructed input given the values of the
191 hidden layer
192
193 """
194 return T.nnet.sigmoid(T.dot(hidden, self.W_prime) + self.b_prime)
195
196 def get_cost_updates(self, contraction_level, learning_rate):
197 """ This function computes the cost and the updates for one trainng

Callers 1

get_cost_updatesMethod · 0.95

Calls

no outgoing calls

Tested by

no test coverage detected