HPOflow Documentation
Tools for Optuna, MLflow and the integration of both.
OptunaMLflow
The main part of this package is the OptunaMLflow class.
It is used as a decorator for Optuna objective functions. If it is applied the Optuna
Study object is augmented.
This augmentation entails writing information to Optuna and MLflow in parallel.
Read more on the OptunaMLflow documentation page.
OptunaMLflowCallback
The OptunaMLflowCallback class integrates
Optuna and
MLflow
with Transformers.
This is done by using OptunaMLflow internally
and the transformers.TrainerCallback to integrate with Transformers.
Read more on the OptunaMLflowCallback documentation page.
SignificanceRepeatedTrainingPruner
This is an Optuna Pruner which uses statistical significance as
an heuristic for decision-making. It prunes repeated trainings like in a cross validation.
As the test method a t-test is used.
Read more on the SignificanceRepeatedTrainingPruner documentation page.
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