Skip to content

chore(people-and-planet-ai): Retire the geospatial-classification sample - #14813

Open
iennae wants to merge 1 commit into
mainfrom
retire-ppai-geospatial-classification
Open

iennae wants to merge 1 commit into
mainfrom
retire-ppai-geospatial-classification

Conversation

@iennae

@iennae iennae commented Nov 14, 2026

Copy link
Copy Markdown
Contributor

Description

Retires the Coal Plant Predictions sample (people-and-planet-ai/geospatial-classification).

The sample no longer runs, so it isn't useful to developers, and keeping it means continuing to update its dependencies:

  • The notebook's setup step installs with -c constraints.txt, a file that was removed in 2023, so the notebook stops before any code runs.
  • task.py uses tf.keras.layers.experimental.preprocessing, which Keras 3 removed, and the serving app can't load the model that training writes.
  • It uses the deprecated COPERNICUS/S2 Earth Engine dataset.
  • CI only targets Python 3.14, which tensorflow 2.21 doesn't support, so the sample isn't exercised.
  • Its labels end in 2021, and many of the labeled plants have since closed or switched to natural gas.

This PR removes the code, data, and images, and replaces README.md with a short note that explains the retirement and links to the last version of the code (636f2f5), so existing links to the directory still resolve. The people-and-planet-ai index marks the entry as retired and keeps the video link.

No region tags or published documentation pages reference this sample. After this merges, #14091 can be closed.

Note

Labeled do not merge until the retirement is signed off.

Checklist

Testing

  • I have tested this change on a live environment and verified it works as intended. (Removal only; all links in the new README and the index resolve.)

Compliance & Style


Post-Approval Actions

  • Please merge this PR for me once it is approved

The Coal Plant Predictions sample (people-and-planet-ai/geospatial-classification)
no longer runs and isn't maintained in a usable state:

- The notebook's setup step installs with `-c constraints.txt`, a file that was
  removed in 2023, so the notebook stops before any code runs.
- task.py uses tf.keras.layers.experimental.preprocessing, which Keras 3
  removed, and the serving app can't load the model that the training job
  writes.
- It uses the deprecated COPERNICUS/S2 Earth Engine dataset.
- noxfile_config.py tests only Python 3.14, which tensorflow 2.21 doesn't
  support, so CI never exercises it.
- Its labels end in 2021, and many of the labeled plants have since closed or
  switched to natural gas.

Almost all recent changes to the directory are dependency bumps. Rather than
keep paying that cost for a sample that can't be run, this removes the code,
data, and images, and replaces README.md with a short note that explains the
retirement and links to the last version of the code (commit 636f2f5). The
people-and-planet-ai index marks the entry as retired and keeps the video link.

No region tags or published documentation pages reference this sample.
@iennae
iennae requested a review from a team as a code owner November 14, 2026 03:50
@iennae iennae added the do not merge Indicates a pull request not ready for merge, due to either quality or timing. label Nov 14, 2026
@iennae
iennae requested review from a team as code owners November 14, 2026 03:50
@product-auto-label product-auto-label Bot added samples Issues that are directly related to samples. api: people-and-planet-ai labels Nov 14, 2026

@gemini-code-assist gemini-code-assist Bot left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Code Review

This pull request retires the 'Coal Plant Predictions' (geospatial-classification) sample by deleting its Jupyter notebook, training scripts, serving application, and configuration files. The README files are updated to document the retirement. The feedback suggests a minor grammatical correction in the updated README to use 'doesn't' instead of 'don't' when referring to 'code'.

* **Training the model**: [TensorFlow] in [Vertex AI]
* **Getting predictions**: [TensorFlow] in [Cloud Run]
* The notebook no longer runs: its setup step depends on a file that was removed
in 2023, and its training and serving code don't work with Keras 3.

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

medium

The word "code" is a singular uncountable noun in this context, so "doesn't" should be used instead of "don't".

Suggested change
in 2023, and its training and serving code don't work with Keras 3.
in 2023, and its training and serving code doesn't work with Keras 3.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

api: people-and-planet-ai do not merge Indicates a pull request not ready for merge, due to either quality or timing. samples Issues that are directly related to samples.

Projects

None yet

Development

Successfully merging this pull request may close these issues.

1 participant