Tools for Optuna, MLflow and the integration of both.

Last update: Nov 20, 2022

Overview

HPOflow - Sphinx DOC

Tools for Optuna, MLflow and the integration of both.

Detailed documentation with examples can be found here: Sphinx DOC

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]

Support and Feedback

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

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

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.

Comments

review README.md and CONTRIBUTING.md
Review README.md and CONTRIBUTING.md

is there something missing? maybe compare with optuna and transformers

spelling

idiomatic english

consistency

correctness

links ok?

...

PS: The real documentation is still missing and a know issue.
opened by PhilipMay 12
add typing in optuna_transformers

@twolffpiggott can you please tell me the type of this?

https://github.com/telekom/HPOflow/blob/e2b0943218af419a79ce95e60b67c9a4c2477349/hpoflow/optuna_transformers.py#L47

opened by PhilipMay 6
add `transformers.py`

@twolffpiggott should we add this here or to an other project we open source?

https://github.com/PhilipMay/mltb/blob/master/mltb/integration/transformers.py
enhancement

opened by PhilipMay 6
Create Sphinx documentation page
[x] setup

[x] make GH action

[x] setup page

[x] change styling to telekom style

~~switch to MD~~

[x] add more content

[x] link from README to page

[x] link from pypi to GH page

[x] add impressum

[x] remove strange mouse over image effect

~~add version info~~

documentation
opened by PhilipMay 4
Problems with direct `_imports.check()` call

When the __init__.py imports OMLflowCallback the optuna_transformers.py script is executed. That executes the _imports.check() call which then throws an exception if transformers or mlflow is not installed. But that should be avoided.

See here: https://github.com/telekom/HPOflow/blob/d1cce5cbc2a84634d1484a053286000dda05b681/hpoflow/optuna_transformers.py#L11-L17

The solution would be to put the _imports.check() call into the constructor. But that is not possible because OMLflowCallback inherits from transformers.

The only solution I have is to put OMLflowCallback into an factory function that creates an OMLflowCallback and does the _imports.check() in there.

@twolffpiggott what do you think?
bug

opened by PhilipMay 3
Flake8 ignore list for Black compatibility
Flake8 raises a warning for "E203" when it encounters a Black decision to insert whitespace before : in slicing syntax.

Black's behaviour is more correct here, so my suggestion is to add "E203" to the flake8 config ignore list.

i.e. in setup.cfg:

[flake8] ... extend-ignore = E203
opened by twolffpiggott 3
Simple Example?

I don't understand how to use this package. Could you provide a basic example? I don't understand the import_structure and how it relates to importing the modules? Thanks

opened by jmrichardson 2
WIP prefix in contrib file

Should this

Create Work In Progress [WIP] pull requests only if you need clarification or an explicit review before you can continue your work item.

be more like this

Add a [WIP] prefix on your pull request name if you need clarification or an explicit review before you can continue your work item.

documentation

opened by PhilipMay 2

Releases(0.1.4)

0.1.4(Aug 14, 2022)

Source code(tar.gz)
Source code(zip)
0.1.3(Jul 17, 2022)
add Python 3.10

Source code(tar.gz)
Source code(zip)
0.1.2(Aug 6, 2021)
add all Optuna versions - see https://github.com/telekom/HPOflow/issues/87

Source code(tar.gz)
Source code(zip)
0.1.1(Aug 3, 2021)

hotfix for " fix Optuna issue since new release" #87
Source code(tar.gz)
Source code(zip)
0.1.0(Jul 20, 2021)
First real Release: 0.1.0

Main components are:

OptunaMLflow - Wrapper to log to Optuna and MLflow at the same time.

OptunaMLflowCallback - Integration of Optuna and MLflow for Transformers.

SignificanceRepeatedTrainingPruner - Optuna pruner which uses statistical significance as an heuristic for decision-making.

Source code(tar.gz)
Source code(zip)
0.1.0rc3(Jul 15, 2021)

Source code(tar.gz)
Source code(zip)

Owner

Telekom Open Source Software

published by Deutsche Telekom AG and partner companies

GitHub Repository https://telekom.github.io/HPOflow/

Tools for Optuna, MLflow and the integration of both.

Related tags

Overview

HPOflow - Sphinx DOC

Table of Contents

Maintainers

Installation

Support and Feedback

Reporting Security Vulnerabilities

Contribution

Code of Conduct

Licensing

Comments

Releases(0.1.4)

0.1.4(Aug 14, 2022)

0.1.3(Jul 17, 2022)

0.1.2(Aug 6, 2021)

0.1.1(Aug 3, 2021)

0.1.0(Jul 20, 2021)

First real Release: 0.1.0

0.1.0rc3(Jul 15, 2021)

Owner

Telekom Open Source Software

ArviZ is a Python package for exploratory analysis of Bayesian models

MiniTorch - a diy teaching library for machine learning engineers

It is a forest of random projection trees

Dual Adaptive Sampling for Machine Learning Interatomic potential.

Python package for stacking (machine learning technique)

Evaluate on three different ML model for feature selection using Breast cancer data.

Python 3.6+ toolbox for submitting jobs to Slurm

All-in-one web-based development environment for machine learning

A Python implementation of the Robotics Toolbox for MATLAB

Python/Sage Tool for deriving Scattering Matrices for WDF R-Adaptors

A simple guide to MLOps through ZenML and its various integrations.

Houseprices - Predict sales prices and practice feature engineering, RFs, and gradient boosting

scikit-learn is a python module for machine learning built on top of numpy / scipy

Simplify stop motion animation with machine learning.

Machine-Learning with python (jupyter)

Machine Learning for RC Cars

Uplift modeling and causal inference with machine learning algorithms

Decision tree is the most powerful and popular tool for classification and prediction

Iris-Heroku - Putting a Machine Learning Model into Production with Flask and Heroku

The Fuzzy Labs guide to the universe of open source MLOps