https://github.com/google/nitroml

NitroML is a modular, portable, and scalable model-quality benchmarking framework for Machine Learning and Automated Machine Learning (AutoML) pipelines.
https://github.com/google/nitroml

automl benchmarking kubeflow machine-learning model-quality modular portable python scale tfx

Last synced: 27 days ago
JSON representation

NitroML is a modular, portable, and scalable model-quality benchmarking framework for Machine Learning and Automated Machine Learning (AutoML) pipelines.

• Host: GitHub
• URL: https://github.com/google/nitroml
• Owner: google
• License: apache-2.0
• Archived: true
• Created: 2020-06-09T14:05:01.000Z (about 4 years ago)
• Default Branch: master
• Last Pushed: 2021-03-10T01:29:59.000Z (over 3 years ago)
• Last Synced: 2024-05-08T21:48:37.861Z (about 2 months ago)
• Topics: automl, benchmarking, kubeflow, machine-learning, model-quality, modular, portable, python, scale, tfx
• Language: Jupyter Notebook
• Homepage:
• Size: 2.27 MB
• Stars: 41
• Watchers: 10
• Forks: 6
• Open Issues: 4
• Metadata Files:
  • Readme: README.md
  • Contributing: CONTRIBUTING.md
  • License: LICENSE
  • Code of conduct: CODE_OF_CONDUCT.md

Lists

• awesome-stars - google/nitroml - NitroML is a modular, portable, and scalable model-quality benchmarking framework for Machine Learning and Automated Machine Learning (AutoML) pipelines. (Jupyter Notebook)

README

        
# NitroML

**NitroML** is a framework for benchmarking Automated Machine Learning (AutoML) pipeline model-quality.

Our mission is to provide _model-quality benchmarking tools to accelerate

AutoML research and development_.

More generally, NitroML enables AutoML research teams to iterate more

quickly on their highly-customized, heterogeneous, and multi-stage pipelines. It

offers machine learning model-quality benchmarking best practices

out-of-the-box, curates raw datasets, and leverages [TFX:OSS] to

scale with Cloud-resources. Its benchmark database and analysis tools ensure

that AutoML teams can be data-driven as they modify their systems.

This is not an officially supported Google product.

For more information, see go/nitroml.