tensorflow版目标检测模型训练，控制台输出信息

最新推荐文章于 2024-07-24 17:11:58 发布

Lovefeifei20200119

最新推荐文章于 2024-07-24 17:11:58 发布

阅读量775

点赞数

分类专栏：学习 Python 深度学习 TensorFlow Object DetectionAPI

本文链接：https://blog.csdn.net/qq_28019591/article/details/94727929

版权

学习同时被 3 个专栏收录

31 篇文章 2 订阅

订阅专栏

深度学习

26 篇文章 1 订阅

订阅专栏

Python

20 篇文章 0 订阅

订阅专栏

site-packages\tensorflow\contrib\slim\python\slim\learning.py中

def train()--------->负责程序运行整体流程

...
        logging.info('Starting Session.')
        ...
        logging.info('Starting Queues.')
        ...
        try:
          while not sv.should_stop():
            total_loss, should_stop = train_step_fn(
                sess, train_op, global_step, train_step_kwargs)
            if should_stop:
              logging.info('Stopping Training.')
              sv.request_stop()
              break
        ...
        if logdir and sv.is_chief:
          logging.info('Finished training! Saving model to disk.')
          ...

def train_step()-------->控制台输出训练信息

  if 'should_log' in train_step_kwargs:
    if sess.run(train_step_kwargs['should_log']):
      logging.info('global step %d: loss = %.4f (%.3f sec/step)',
                   np_global_step, total_loss, time_elapsed)