https://github.com/googlecloudplatform/ml-testing-accelerators

Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)
https://github.com/googlecloudplatform/ml-testing-accelerators

gpu machine-learning testing-accelerators tpu

Last synced: 18 days ago
JSON representation

Testing framework for Deep Learning models (Tensorflow and PyTorch) on Google Cloud hardware accelerators (TPU and GPU)

• Host: GitHub
• URL: https://github.com/googlecloudplatform/ml-testing-accelerators
• Owner: GoogleCloudPlatform
• License: apache-2.0
• Created: 2020-03-04T15:21:13.000Z (about 5 years ago)
• Default Branch: master
• Last Pushed: 2024-11-22T20:45:21.000Z (5 months ago)
• Last Synced: 2025-03-30T15:43:17.355Z (23 days ago)
• Topics: gpu, machine-learning, testing-accelerators, tpu
• Language: Jsonnet
• Homepage:
• Size: 3.97 MB
• Stars: 64
• Watchers: 30
• Forks: 60
• Open Issues: 35
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE

Awesome Lists containing this project

README

        
> **_IMPORTANT:_**  This repository is being deprecated. Please migrate or onboard your ML tests to our new repository [here](https://github.com/GoogleCloudPlatform/ml-auto-solutions).

# ML Testing Accelerators

A set of tools and examples to run machine learning tests on ML hardware

accelerators (TPUs or GPUs) using Google Cloud Platform.

This is not an officially supported Google product.

## Getting Started

In this mode, your tests and/or models run on an automated schedule in GKE.

Results are collected by the "Metrics Handler" and written to BigQuery.

1. Install all of our [development prerequisites](doc/developing.md#Prerequisites).

1. Follow instructions in the [`deployments`](deployments/README.md) directory to set up a Kubernetes Cluster.

1. Follow instructions in the [`images`](images/README.md) directory to set up the Docker image that your tests will run.

1. Deploy the [metrics handler](metrics/README.md#metrics-handler) to [Google Cloud Functions](https://cloud.google.com/functions).

1. Deploy the [event publisher](metrics/README.md#event-publisher) to you GKE cluster.

1. See [`templates`](templates/README.md) directory for a [JSonnet](https://jsonnet.org/) template library to generate test config files.

1. (Optional) Set up a dashboard to view test results. See [ dashboard ](dashboard/README.md) directory for instructions.

Are you interested in using ML Testing Accelerators? E-mail [[email protected]](mailto:[email protected]) and tell us about your use-case. We're happy to help you get started.