External library integrations for Ray Tune#

Tune Experiment Monitoring Integrations#

Comet (air.integrations.comet)#

See here for an example.

CometLoggerCallback([online, tags, ...])

CometLoggerCallback for logging Tune results to Comet.

MLflow (air.integrations.mlflow)#

See here for an example.

MLflowLoggerCallback([tracking_uri, ...])

MLflow Logger to automatically log Tune results and config to MLflow.

setup_mlflow([config, tracking_uri, ...])

Set up a MLflow session.

Weights and Biases (air.integrations.wandb)#

See here for an example.

WandbLoggerCallback([project, group, ...])

Weights and biases (https://www.wandb.ai/) is a tool for experiment tracking, model optimization, and dataset versioning.

setup_wandb([config, api_key, api_key_file, ...])

Set up a Weights & Biases session.

Integrations with ML Libraries#

Keras (air.integrations.keras)#

ReportCheckpointCallback([checkpoint_on, ...])

Keras callback for Ray AIR reporting and checkpointing.

MXNet (tune.integration.mxnet)#

TuneReportCallback([metrics])

MXNet to Ray Tune reporting callback

TuneCheckpointCallback([filename, frequency])

MXNet checkpoint callback

PyTorch Lightning (tune.integration.pytorch_lightning)#

TuneReportCallback([metrics, on])

PyTorch Lightning to Ray Tune reporting callback

TuneReportCheckpointCallback([metrics, ...])

PyTorch Lightning report and checkpoint callback

XGBoost (tune.integration.xgboost)#

TuneReportCallback([metrics, ...])

XGBoost to Ray Tune reporting callback

TuneReportCheckpointCallback([metrics, ...])

XGBoost report and checkpoint callback

LightGBM (tune.integration.lightgbm)#

TuneReportCallback([metrics, ...])

Create a callback that reports metrics to Ray Tune.

TuneReportCheckpointCallback([metrics, ...])

Creates a callback that reports metrics and checkpoints model.