Make autoencoder easier to copy-paste

[beyarkay]

This PR simply removes the explicit (28, 28) and 784 which represented the shape of the FMNIST dataset.

It rather uses a shape member variable which is passed into the constructor for the Autoencoder. This shape variable is set to be shape = x_test.shape[1:] by default.

I often come to this page, since I forget exactly how to structure an Autoencoder in TF, and I always end up writing this little snippet out where I extract the shape variable into a member variable of the Autoencoder class.

[github-actions]

Preview

Preview and run these notebook edits with Google Colab:

• site/en/tutorials/generative/autoencoder.ipynb

Rendered notebook diffs available on ReviewNB.com.

Format and style

Use the TensorFlow docs notebook tools to format for consistent source diffs and lint for style:

$ python3 -m pip install -U --user git+https://github.com/tensorflow/docs

$ python3 -m tensorflow_docs.tools.nbfmt notebook.ipynb

$ python3 -m tensorflow_docs.tools.nblint --arg=repo:tensorflow/docs notebook.ipynb

If commits are added to the pull request, synchronize your local branch: git pull origin patch-1

[beyarkay]

Just wanting to check in, and see if this sort of PR is welcome here?

[vijayasarathib]

LGTM

[markmcd]

Thanks for the submission - this is great.

[markmcd]

Smoke tests found an issue.

[markmcd]

👍

[beyarkay]

Thanks for the approval! How often are the docs re-built?

[markmcd]

This one is updated nightly, but before it's merged it requires another internal review, so could take a day or two.

[beyarkay]

Thanks for helping out!