Question

在安装了zeppelin，Anaconda和tensorflow的服务器上运行良好。但是当我想在齐柏林飞艇上划下代码时如下：

%python
import pandas as pd
import tensorflow as tf

并运行此代码，我遇到了这样的错误：

Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/__init__.py", line 24, in <module>
    from tensorflow.python import *
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/__init__.py", line 71, in <module>
    from tensorflow.python.framework.framework_lib import *
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/framework_lib.py", line 73, in <module>
    from tensorflow.python.framework.ops import Graph
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/ops.py", line 41, in <module>
    from tensorflow.python.framework import registry
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/framework/registry.py", line 28, in <module>
    from tensorflow.python.platform import tf_logging as logging
  File "/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/platform/tf_logging.py", line 53, in <module>
    if _interactive:
NameError: name '_interactive' is not defined

如何解决这个问题。我没有谷歌这个解决方案。

Answer 1

最后，我修复了文件

/home/hadoop/anaconda3/lib/python3.5/site-packages/tensorflow/python/platform/tf_logging.py

像这样：

# If we are in an interactive environment (like jupyter), set loglevel to info
# and pipe the output to stdout
if True:
#if _interactive: 
  _logger.setLevel(INFO)
  _logging_target = _sys.stdout
else:
  _logging_target = _sys.stderr

并重新启动zeppelin的python解释器。它工作正常！

Answer 2

tensorflow项目中有一个问题：https://github.com/tensorflow/tensorflow/pull/8420

他们通过使用默认值False

实现_interactive来修复它

# If we are in an interactive environment (like jupyter), set loglevel to info
# and pipe the output to stdout
_interactive = False
if _interactive: 
  _logger.setLevel(INFO)
  _logging_target = _sys.stdout
else:
  _logging_target = _sys.stderr

最好是在错误修复程序包含在发行版中时立即更新tensorflow，或者像上面提到的提交中那样更新/site-packages/tensorflow/python/platform/tf_logging.py。

此致卢瓦克

tensorflow无法与zeppelin一起使用

2 个答案: