ModuleNotFoundError: No module named 'tensorflow'

Posted in Python by Dirk - last update: Feb 02, 2024

Python raises the ModuleNotFoundError: No module named 'tensorflow when it is unable to find the tensorflow library. The most likely cause is that you didn’t install tensorflow in the environment where you are running your code. Quick fix: install tensorflow using: the pip install tensorflow command

Possible causes and solutions

TensorFlow is an open-source machine learning (ML) framework developed by Google Brain. It provides a comprehensive set of tools and libraries for building and deploying machine learning models.

import tensorflow as tf
from tensorflow.keras import layers, models
from tensorflow.keras.datasets import mnist
from tensorflow.keras.utils import to_categorical

We have all been there: the dreaded ’no module named error’ - you try running your code, and there it is:

Traceback (most recent call last):
  File "myfile.py", line 1, in <module>
ModuleNotFoundError: No module named 'tensorflow'

Luckily, it is easy to fix. Below you can find the most common causes and how to fix them:

’tensorflow’ Library is not installed

In your code you are importing the tensorflow module

import tensorflow

which generates the ModuleNotFoundError: No module named 'tensorflow’. The the most common cause for this error: the environment where you are running your code doesn’t have the tensorflow library installed.

You can use the following command in the terminal or command prompt to check if the tensorflow library is installed:

pip show tensorflow

If tensorflow is not installed you will get this result:

WARNING: Package(s) not found: tensorflow

Solution:

pip install tensorflow

’tensorflow’ Module installed in the Wrong Directory

It is possible that you did install the tensorflow module, but that the directory where you installed it is not in the Python path.

You can check the Python path in a few different ways:

1. Using sys module in Python:

Open a Python script or the Python interpreter and run the following code:

import sys
print(sys.path)

This will print a list of directories that make up the Python path.

2. Using Command Line

Open a command prompt or terminal and run:

Windows

echo %PYTHONPATH%

Unix/Linux/macOS

echo $PYTHONPATH

If the PYTHONPATH environment variable is set, it will display the directories in the Python path.

Solution: Move the tensorflow module to a directory that is included in the Python path or add the tensorflow module’s directory to the sys.path using sys.path.append().

’tensorflow’ not installed in Virtual Environment

Even if you have tensorflow installed in on your computer, if you are using a virtual environment you cannot access it from this environment.

If you are using virtual environments, ensure that the virtual environment is activated, and that the tensorflow modules is installed within this environment.

Installation is simple:

pip install tensorflow

File Naming Conflicts:

It is not really likely - but if you happen to have a tensorflow.py file in your environment and you also installed the standard tensorflow library it will cause a conflict. Rename your script/module to avoid conflicts with existing module names.