Skip to content

Instantly share code, notes, and snippets.

Embed
What would you like to do?
positive_gradient_matrix = tf.multiply(input_matrix, self._hidden_states)
negative_gradient_matrix = tf.multiply(self._visible_cdstates, self._hidden_cdstates)
new_weights = self._weights
new_weights.assign_add(tf.multiply(positive_gradient_matrix, self._leraning_rate))
new_weights.assign_sub(tf.multiply(negative_gradient_matrix, self._leraning_rate))
self._training = tf.assign(self._weights, new_weights)
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment