DOCSEARCHbyAlgolia
Submit Docs

Google Cloud Platform Python Documentation Samples

repository·main·Indexed 27 days ago

https://github.com/googlecloudplatform/python-docs-samples

A repository of Python code samples for various Google Cloud Platform products, designed to help developers use Google Cloud APIs and client libraries. Includes examples for App Engine Flexible Environment (Django, Flask, SciPy, Cloud SQL, Cloud Storage, Pub/Sub), Anti Money Laundering AI (AML AI), and deploying an MCP server on Cloud Run.

Tokens
217.1K
Snippets
722
Records
1.2K
Agent score
93%

What's inside python-docs-samples

  2. Overview of Global Fishing Watch Timeseries Classification

    main

    This project implements a machine learning pipeline to classify whether a ship is fishing or not on an hourly basis using Maritime Mobile Service Identity (MMSI) location data. The solution utilizes a 1D Fully Convolutional Network built with Keras.

    The pipeline architecture consists of:

    • Dataset Creation: Using Apache Beam on Google Cloud Dataflow.
    • Model Training: Using Keras on Vertex AI.
    • Inference/Predictions: Using Keras deployed on Cloud Run.

  3. Overview of Cloud Run functions versions

    main

    This repository contains samples for two versions of Cloud Run functions:

    • Cloud Run functions (2nd gen): Formerly known as Cloud Functions (2nd gen). These deploy as services on Cloud Run and can be triggered using Eventarc and Pub/Sub. They are created using gcloud functions or gcloud run. Samples are located in the functions/v2/ folder.
    • Cloud Run functions (1st gen): Formerly known as Cloud Functions (1st gen). These have limited event triggers and configurability. They are created using gcloud functions --no-gen2. Samples are located in the current functions/ folder.

  8. Overview of Weather Forecasting with Timeseries Regression

    main

    This project demonstrates a weather forecasting model that uses satellite data to predict precipitation for the next 2 and 6 hours. The workflow involves:

    • Model Architecture: A 2D Fully Convolutional Network implemented in PyTorch.
    • Dataset Creation: Uses Sentinel-2 satellite data and ESA WorldCover sourced from Google Earth Engine, processed using Dataflow.
    • Model Training: Performed using PyTorch on Vertex AI.
    • Inference: Predictions can be generated using PyTorch in a local environment.

  9. Overview of Coal Plant Predictions using Geospatial Classification

    main

    This project demonstrates a machine learning workflow that uses satellite data to predict whether a coal plant is operational and producing carbon emissions.

    Key components of the pipeline include:

    • Data Source: Sentinel-2 satellite data retrieved from Google Earth Engine.
    • Model Architecture: A 1D Fully Convolutional Network implemented in TensorFlow.
    • Training: Model training is performed using TensorFlow on Vertex AI.
    • Inference: Predictions are served using TensorFlow on Cloud Run.

  12. Overview of Dataflow custom containers

    main
    This repository contains samples demonstrating common use cases for creating custom container images for Dataflow workers. Custom containers allow you to specify the environment, dependencies, and system libraries required by your Dataflow pipeline. For detailed official documentation on implementation, refer to the Using custom containers in Dataflow guide.