尽管Tensorboard的事件太大，但Tensorboard的步骤太少了

Question

我已经粘贴了代码的一部分（所有的部分是指张力索板部分）如下。我只记录了损失标量变量，只为一个epoch添加一次摘要。我一共运行了3个时期。理想情况下应该是一个非常小的tfevents文件。但是，tfevents文件是1.3GB。我不知道是什么原因导致文件这么大。 如果需要的话，很乐意分享剩下的代码

def do_training(update_op, loss, summary_op):

    writer = tf.summary.FileWriter(logs_path, graph=tf.get_default_graph())
    saver = tf.train.Saver()
    with tf.Session() as sess:
        sess.run(tf.global_variables_initializer())
        try:
            step = 0
            while True:

                if step % (X_train.shape[0]/batch_size) == 0:
                    _, loss_value = sess.run((update_op, loss))
                    summary = sess.run(summary_op)
                    writer.add_summary(summary, global_step=step)
                    print('Step {} with loss {}'.format(step, loss_value))
                else:
                    _, loss_value = sess.run((update_op, loss))
                step += 1

        except tf.errors.OutOfRangeError:
            # we're through the dataset
            pass
        writer.close()
        saver.save(sess,save_path)
    print('Final loss: {}'.format(loss_value))

def serial_training(model_fn, dataset):

    iterator = dataset.make_one_shot_iterator()

    loss = model_fn(lambda: iterator.get_next())
    tf.summary.scalar("loss", loss)
    summary_op = tf.summary.merge_all()
    optimizer = tf.train.AdamOptimizer(learning_rate=0.0002)
    global_step = tf.train.get_or_create_global_step()


    update_op = optimizer.minimize(loss, global_step=global_step)

    do_training(update_op, loss, summary_op)

tf.reset_default_graph()

serial_training(training_model,training_dataset(epochs=3,batch_size=batch_size))