Method train

a02_TextCNN/p7_TextCNN_model.py:191–201

based on the loss, use SGD to update parameter

(self)

        return train_op

    def train(self):
        """based on the loss, use SGD to update parameter"""
        learning_rate = tf.train.exponential_decay(self.learning_rate, self.global_step, self.decay_steps, self.decay_rate, staircase=True)
        self.learning_rate_=learning_rate
        optimizer = tf.train.AdamOptimizer(learning_rate)
        gradients, variables = zip(*optimizer.compute_gradients(self.loss_val))
        gradients, _ = tf.clip_by_global_norm(gradients, 5.0)
        update_ops = tf.get_collection(tf.GraphKeys.UPDATE_OPS) #ADD 2018.06.01
        with tf.control_dependencies(update_ops):  #ADD 2018.06.01
            train_op = optimizer.apply_gradients(zip(gradients, variables))
        return train_op

#test started. toy task: given a sequence of data. compute it's label: sum of its previous element,itself and next element greater than a threshold, it's label is 1,otherwise 0.
#e.g. given inputs:[1,0,1,1,0]; outputs:[0,1,1,1,0].