ImportError: No module named tensorflow

问题:

Please help me with this error

I have installed the tensorflow module on my server and below is it's information

15IT60R19@cpusrv-gpu-109:~$ pip show tensorflow
Name: tensorflow
Version: 1.0.0
Summary: TensorFlow helps the tensors flow
Home-page: http://tensorflow.org/
Author: Google Inc.
Author-email: [e-mail]
License: Apache 2.0
Location: /home/other/15IT60R19/anaconda2/lib/python2.7/site-packages
Requires: mock, numpy, protobuf, wheel, six

But when I try to import tensorflow I get following error

>>> import tensorflow as tf
Traceback (most recent call last):
  File "<stdin>", line 1, in <module>
ImportError: No module named tensorflow

My python version is as following

Python 2.7.12 |Anaconda 2.5.0 (64-bit)| (default, Jul  2 2016, 17:42:40) 
[GCC 4.4.7 20120313 (Red Hat 4.4.7-1)] on linux2
Type "help", "copyright", "credits" or "license" for more information.
Anaconda is brought to you by Continuum Analytics.
Please check out: http://continuum.io/thanks and https://anaconda.org

I have tried the solutions given in sol1

Sol2

I do not have sudo access to the server I can only use pip to install any module

回答1:

Try installing tensorflow again with the whatever version you want and with option --ignore-installed like:

pip install tensorflow==1.2.0 --ignore-installed

I solved same issue using this command.

回答2:

I had a more basic problem when I received this error.

The "Validate your installation" instructions say to type: python

However, I have both 2.7 and 3.6 installed. Because I used pip3 to install tensorflow, I needed to type: python3

Using the correct version, I could import the "tensorflow" module.

回答3:

Check if Tensorflow was installed successfully using:

 pip3 show tensorflow

If you get something like

Name: tensorflow
Version: 1.2.1
Summary: TensorFlow helps the tensors flow
Home-page: http://tensorflow.org/
Author: Google Inc.
Author-email: [e-mail]
License: Apache 2.0
Location: /usr/local/lib/python3.5/dist-packages
Requires: bleach, markdown, html5lib, backports.weakref, werkzeug, numpy, protobuf, wheel, six

You may try adding the path of your tensorflow location by:

export PYTHONPATH=/your/tensorflow/path:$PYTHONPATH.

回答4:

Try installing tensorflow in the user site - This installation only works for you.

pip install tensorflow --user

回答5:

For Anaconda3, simply install in Anaconda Navigator:

回答6:

You may need this since first one may not work.

python3 -m pip install --upgrade https://storage.googleapis.com/tensorflow/mac/cpu/tensorflow-0.12.0-py3-none-any.whl

回答7:

I was trying to install tensorflow GPU for a Windows 7 with pip3 for python3.5.x. Instead of doing pip3 install --upgrade tensorflow I just did pip install tensorflow and after it completed i was finally able to import tensorflow in python3.5.x.

回答8:

Instead of using the doc's command (conda create -n tensorflow pip python=2.7 # or python=3.3, etc.) which wanted to install python2.7 in the conda environment, and kept erroring out saying the module can't be found when following the installation validation steps, I used conda create -n tensorflow pip python=3 to make sure python3 was installed in the environment.

Doing this, I only had to type python instead of python3 when validating the installation and the error went away.

回答9:

Try Anaconda install steps from TensorFlow docs.

回答10:

Activate the virtualenv environment by issuing one of the following commands:

$ source ~/tensorflow/bin/activate # bash, sh, ksh, or zsh
$ source ~/tensorflow/bin/activate.csh # csh or tcsh

Hope this help

回答11:

This Worked for me:

$ sudo easy_install pip
$ sudo easy_install --upgrade six
$ export TF_BINARY_URL=https://storage.googleapis.com/tensorflow/mac/tensorflow-0.9.0-py2-none-any.whl
$ sudo pip install --upgrade $TF_BINARY_URL

回答12:

with python2

pip show tensorflow to check install

python test.py to run test

with python3

pip3 show tensorflow to check install

python3 test.py to run test

test.py

import tensorflow as tf
import numpy as np

c = np.array([[3.,4], [5.,6], [6.,7]])
step = tf.reduce_mean(c, 1)                                                                                 
with tf.Session() as sess:
    print(sess.run(step))

Or, if you haven't install tensorflow yet, try the offical document

回答13:

you might wanna try this:

$conda install -c conda-forge tensorflow