Question

我设置了一个大的纪元号，我想提前停止它。简单地杀死python进程似乎不起作用（我试过一个训练过程）。这是我在训练期间或被杀之后使用诸如summary_iterator 之类的工具阅读摘要文件时得到的结果：

Traceback (most recent call last):
  File "tensorboard_reader.py", line 47, in <module>
    train_loss = reader(train_file[0], ['cost', 'likelihood_loss', 'focal_loss'])
  File "tensorboard_reader.py", line 13, in reader
    for e in tf.train.summary_iterator(filename):
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/summary/summary_iterator.py", line 353, in summary_iterator
    for r in tf_record.tf_record_iterator(path):
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/lib/io/tf_record.py", line 77, in tf_record_iterator
    reader.GetNext(status)
  File "/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/errors_impl.py", line 473, in __exit__
    c_api.TF_GetCode(self.status.status))
tensorflow.python.framework.errors_impl.DataLossError: truncated record at 27159094

我可以停止tensorflow培训，还能安全地关闭汇总作者的tfrecords文件吗？

0 个答案: