Cannot evaluate tensor using 'eval' No default session is registered错误问题的解决

发布于 2019-09-26 作者 风铃 87次 浏览 版块 前端

原文:https://blog.csdn.net/dongdong9223/article/details/78968824

Python如下代码:

#coding=utf-8

import tensorflow as tf
# NumPy是一个科学计算的工具包,这里通过NumPy工具包生成模拟数据集,
from numpy.random import RandomState

v = tf.constant([[1.0, 2.0, 3.0], [4.0, 5.0, 6.0]])

'''
#这段代码之前如果没有添加session,会报错:Cannot evaluate tensor using eval(): No default session is registered
print tf.clip_by_value(v, 2.5, 4.5).eval()

为避免报错应该使用下面的方式:
with tf.Session() as sess:
print tf.clip_by_value(v, 2.5, 4.5).eval()

'''

with tf.Session() as sess:
print tf.clip_by_value(v, 2.5, 4.5).eval()

  • 会报如下错误:

Traceback (most recent call last): 
File “constant_eval.py”, line 10, in 
print tf.clip_by_value(v, 2.5, 4.5).eval() 
File “/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py”, line 575, in eval 
return _eval_using_default_session(self, feed_dict, self.graph, session) 
File “/usr/local/lib/python2.7/dist-packages/tensorflow/python/framework/ops.py”, line 3619, in _eval_using_default_session 
raise ValueError(“Cannot evaluate tensor using eval(): No default ” 
ValueError: Cannot evaluate tensor using eval(): No default session is registered. Use with sess.as_default()or pass an explicit session to eval(session=sess)

如图所示:

这里写图片描述

可以如代码所示增加session的定义:

with tf.Session() as sess:
  • 1

让print在其中运行就可以了,如图所示:

这里写图片描述

收藏
暂无回复