Learn how to train machine learning models for classification and prediction by following the steps in
interactive notebooks. These tutorials integrate Dataflow into
end-to-end machine learning workflows. You can also view the tutorials in
[GitHub](https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai).

*** ** * ** ***

## Land cover image segmentation

[Video](https://www.youtube.com/watch?v=zImQf91ffFo)

This land classification model uses a [TensorFlow](https://www.tensorflow.org/)
framework and satellite data from
[Google Earth Engine](https://earthengine.google.com/) to demonstrate semantic segmentation.
The tutorial uses [TensorFlow in Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/machine-learning/start/tensorflow)
to train the model, TensorFlow in [Cloud Run](https://docs.cloud.google.com/run/docs) to
make real-time predictions, and Dataflow to make batch predictions.
[View the code on GitHub.](https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai/land-cover-classification)

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.sandbox.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/land-cover-classification/README.ipynb)

*** ** * ** ***

## Weather forecasting time series regression

[Video](https://www.youtube.com/watch?v=6-UJzEXMvGY)

This weather forecasting model uses a [PyTorch](https://docs.cloud.google.com/vertex-ai/docs/start/pytorch)
framework and satellite data from
[Google Earth Engine](https://earthengine.google.com/) to
forecast precipitation for the next two and six hours.
The tutorial uses PyTorch to create a fully convolutional network,
[Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/machine-learning) to train the
model, Dataflow to create the dataset, and PyTorch to make local predictions.
[View the code on GitHub.](https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai/weather-forecasting)

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.sandbox.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/weather-forecasting/notebooks/1-overview.ipynb)

*** ** * ** ***

## Global fishing watch time series classification

[Video](https://www.youtube.com/watch?v=LnEhSVEJUuY)

This classification model uses a [TensorFlow](https://www.tensorflow.org/)
framework and Maritime Mobile Service Identity
(MMSI) location data to classify whether a ship is fishing every hour.
The tutorial uses [Keras](https://keras.io/) and TensorFlow to train the
model, Dataflow to create the dataset, and Keras in
[Cloud Run](https://docs.cloud.google.com/run/docs) to make local predictions.
[View the code on GitHub.](https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai/timeseries-classification)

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.sandbox.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/timeseries-classification/README.ipynb)

*** ** * ** ***

## Wildlife image classification

[Video](https://www.youtube.com/watch?v=hUzODH3uGg0)

This classification model uses an [AutoML](https://docs.cloud.google.com/vertex-ai/docs) framework to
create a model trained to recognize animal species from camera trap pictures.
The tutorial uses [AutoML in Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/machine-learning/beginner/beginners-guide)
to train the model, Dataflow to create the dataset, and
[Gemini Enterprise Agent Platform](https://docs.cloud.google.com/gemini-enterprise-agent-platform/machine-learning) to make predictions.
[View the code on GitHub.](https://github.com/GoogleCloudPlatform/python-docs-samples/tree/main/people-and-planet-ai/image-classification)

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.sandbox.google.com/github/GoogleCloudPlatform/python-docs-samples/blob/main/people-and-planet-ai/image-classification/README.ipynb)