Ray Tune API

The Ray Tune API reference documents Tune’s public Python API and CLI. Each page below covers one group of APIs.

• Tune Execution (tune.Tuner)
  • Tuner
  • tune.run_experiments
• Tune Experiment Results (tune.ResultGrid)
  • ResultGrid (tune.ResultGrid)
  • Result (tune.Result)
  • ExperimentAnalysis (tune.ExperimentAnalysis)
• Training in Tune (tune.Trainable, tune.report)
  • Function Trainable API
  • Class Trainable API
  • Comparing Tune’s Function API and Class API
  • Advanced Resource Allocation
  • Function API
  • Trainable (Class API)
  • Tune Trainable Utilities
• Tune Search Space API
  • Random Distributions API
  • Grid Search and Custom Function APIs
  • References
• Tune Search Algorithms (tune.search)
  • Saving and Restoring Tune Search Algorithms
  • Random search and grid search (tune.search.basic_variant.BasicVariantGenerator)
  • Ax (tune.search.ax.AxSearch)
  • Bayesian Optimization (tune.search.bayesopt.BayesOptSearch)
  • BOHB (tune.search.bohb.TuneBOHB)
  • HEBO (tune.search.hebo.HEBOSearch)
  • HyperOpt (tune.search.hyperopt.HyperOptSearch)
  • Nevergrad (tune.search.nevergrad.NevergradSearch)
  • Optuna (tune.search.optuna.OptunaSearch)
  • ZOOpt (tune.search.zoopt.ZOOptSearch)
  • Repeated Evaluations (tune.search.Repeater)
  • ConcurrencyLimiter (tune.search.ConcurrencyLimiter)
  • Custom Search Algorithms (tune.search.Searcher)
  • Shim Instantiation (tune.create_searcher)
• Tune Trial Schedulers (tune.schedulers)
  • ASHA (tune.schedulers.ASHAScheduler)
  • HyperBand (tune.schedulers.HyperBandScheduler)
  • Median Stopping Rule (tune.schedulers.MedianStoppingRule)
  • Population Based Training (tune.schedulers.PopulationBasedTraining)
  • Population Based Training Replay (tune.schedulers.PopulationBasedTrainingReplay)
  • Population Based Bandits (PB2) (tune.schedulers.pb2.PB2)
  • BOHB (tune.schedulers.HyperBandForBOHB)
  • ResourceChangingScheduler
  • FIFOScheduler (Default Scheduler)
  • TrialScheduler Interface
  • Shim Instantiation (tune.create_scheduler)
• Tune Stopping Mechanisms (tune.stopper)
  • Stopper Interface (tune.Stopper)
  • Tune Built-in Stoppers
• Tune Console Output (Reporters)
  • Reporter Interface (tune.ProgressReporter)
  • Tune Built-in Reporters
• Syncing in Tune
  • Tune Syncing Configuration
• Tune Loggers (tune.logger)
  • LoggerCallback Interface (tune.logger.LoggerCallback)
  • Tune Built-in Loggers
  • MLFlow Integration
  • Wandb Integration
  • Comet Integration
  • Aim Integration
  • Other Integrations
• Tune Callbacks (tune.Callback)
  • Callback Interface
• Environment variables used by Ray Tune
• External library integrations for Ray Tune
  • PyTorch Lightning (tune.integration.pytorch_lightning)
  • XGBoost (tune.integration.xgboost)
  • LightGBM (tune.integration.lightgbm)
• Tune Internals
  • TunerInternal
  • Trial
  • FunctionTrainable
  • Registry
  • Output
• Tune CLI (Experimental)