tensorflow: Module Not Found in docker build: tensorflow_datasets

I have been following the Docker instructions on the Tensorflow website (https://www.tensorflow.org/install/docker). Trying to run the Jupyter notebook provided in the tensorflow docker images, specifically trying to run the tutorial here: https://github.com/tensorflow/docs/blob/master/site/en/tutorials/keras/text_classification.ipynb

But when I run the code segments in sequence, I get the following error:

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
<ipython-input-3-7eaaf952f6e2> in <module>
      1 from tensorflow import keras
      2 
----> 3 import tensorflow_datasets as tfds
      4 tfds.disable_progress_bar()
      5 

ModuleNotFoundError: No module named 'tensorflow_datasets'

This would seem to be an issue with the Docker image since I would expect that module to be installed by default.

System information

OS Platform and Distribution: MacOS 10.15
TensorFlow installed from (source or binary): docker run -it -p 8888:8888 tensorflow/tensorflow:nightly-py3-jupyter

About this issue

Original URL
State: closed
Created 5 years ago
Comments: 16 (4 by maintainers)

Most upvoted comments

Inside of the notebook you can run !pip install tensorflow_datasets before the code that imports tensorflow_datasets. Similar to how !pip install seaborn is done in regression.ipynb.

That change could be submitted as a pull request to https://github.com/tensorflow/docs

wdirons on Oct 21, 2019