HPOflow - Sphinx DOC

We decided to archive this project and migrate the most important functionality to MLTB2.

Tools for Optuna, MLflow and the integration of both.

Detailed documentation with examples can be found here: Sphinx DOC

Table of Contents

• Maintainers
• Installation
• Support and Feedback
• Reporting Security Vulnerabilities
• Contribution
• Code of Conduct
• Licensing

Maintainers

This project is maintained by the One Conversation team of Deutsche Telekom AG.

The main components are:

• hpoflow.OptunaMLflow:
  A wrapper to use Optuna and log to MLflow at the same time.
• hpoflow.OptunaMLflowCallback:
  Class inheriting from transformers.TrainerCallback that integrates with OptunaMLflow to send the logs to MLflow and Optuna during model training.
• hpoflow.SignificanceRepeatedTrainingPruner:
  An Optuna pruner to use statistical significance (a t-test which serves as a heuristic) to stop unpromising trials early, avoiding unnecessary repeated training during cross validation.

Installation

HPOflow is available at the Python Package Index (PyPI). It can be installed with pip:

$ pip install hpoflow

Some additional dependencies might be necessary.

To use hpoflow.optuna_mlflow.OptunaMLflow:

$ pip install mlflow GitPython

To use hpoflow.optuna_transformers.OptunaMLflowCallback:

$ pip install mlflow GitPython transformers

To install all optional dependencies use:

$ pip install hpoflow[optional]

To install all dependencies use:

$ pip install hpoflow[all]

Release News

Here you can find the latest versions of the software:

• pypi package
• GitHub releases

Important news and features in the releases:

• add Python 3.10 support and remove Python 3.6 support #95 - version 0.1.4 at 2022-08-14

Support and Feedback

The following channels are available for discussions, feedback, and support requests:

• open an issue in our GitHub repository
• send an e-mail to our open source team

Reporting Security Vulnerabilities

This project is built with security and data privacy in mind to ensure your data is safe. We are grateful for security researchers and users reporting a vulnerability to us, first. To ensure that your request is handled in a timely manner and non-disclosure of vulnerabilities can be assured, please follow the below guideline.

Please do not report security vulnerabilities directly on GitHub. GitHub Issues can be publicly seen and therefore would result in a direct disclosure.

Please address questions about data privacy, security concepts, and other media requests to the [email protected] mailbox.

Contribution

Our commitment to open source means that we are enabling - in fact encouraging - all interested parties to contribute and become part of our developer community.

Contribution and feedback is encouraged and always welcome. For more information about how to contribute, as well as additional contribution information, see our Contribution Guidelines.

Code of Conduct

This project has adopted the Contributor Covenant as our code of conduct. Please see the details in our Contributor Covenant Code of Conduct. All contributors must abide by the code of conduct.

Licensing

Copyright (c) 2021 Philip May, Deutsche Telekom AG
Copyright (c) 2021 Philip May
Copyright (c) 2021 Timothy Wolff-Piggott

Licensed under the MIT License (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License by reviewing the file LICENSE in the repository.