seq2seq_model.py 文件源码

def init_optimizer(self):
        print("setting optimizer..")
        # Gradients and SGD update operation for training the model
        trainable_params = tf.trainable_variables()
        if self.optimizer.lower() == 'adadelta':
            self.opt = tf.train.AdadeltaOptimizer(learning_rate=self.learning_rate)
        elif self.optimizer.lower() == 'adam':
            self.opt = tf.train.AdamOptimizer(learning_rate=self.learning_rate)
        elif self.optimizer.lower() == 'rmsprop':
            self.opt = tf.train.RMSPropOptimizer(learning_rate=self.learning_rate)
        else:
            self.opt = tf.train.GradientDescentOptimizer(learning_rate=self.learning_rate)

        # Compute gradients of loss w.r.t. all trainable variables
        gradients = tf.gradients(self.loss, trainable_params)

        # Clip gradients by a given maximum_gradient_norm
        clip_gradients, _ = tf.clip_by_global_norm(gradients, self.max_gradient_norm)

        # Update the model
        self.updates = self.opt.apply_gradients(
            zip(clip_gradients, trainable_params), global_step=self.global_step)