API Reference

• Algorithms
  • Base
  • Q-learning
  • Decision Transformer
• Q Functions
  • d3rlpy.models.MeanQFunctionFactory
  • d3rlpy.models.QRQFunctionFactory
  • d3rlpy.models.IQNQFunctionFactory
• Replay Buffer
  • MDPDataset
  • Replay Buffer
  • Buffer
  • TransitionPicker
  • TrajectorySlicer
  • WriterPreprocess
• Datasets
  • d3rlpy.datasets.get_cartpole
  • d3rlpy.datasets.get_pendulum
  • d3rlpy.datasets.get_atari
  • d3rlpy.datasets.get_atari_transitions
  • d3rlpy.datasets.get_d4rl
  • d3rlpy.datasets.get_dataset
  • d3rlpy.datasets.get_minari
• Preprocessing
  • Observation
  • Action
  • Reward
• Optimizers
  • d3rlpy.models.OptimizerFactory
  • d3rlpy.models.SGDFactory
  • d3rlpy.models.AdamFactory
  • d3rlpy.models.RMSpropFactory
  • d3rlpy.models.GPTAdamWFactory
• Network Architectures
  • d3rlpy.models.DefaultEncoderFactory
  • d3rlpy.models.PixelEncoderFactory
  • d3rlpy.models.VectorEncoderFactory
• Metrics
  • d3rlpy.metrics.TDErrorEvaluator
  • d3rlpy.metrics.DiscountedSumOfAdvantageEvaluator
  • d3rlpy.metrics.AverageValueEstimationEvaluator
  • d3rlpy.metrics.InitialStateValueEstimationEvaluator
  • d3rlpy.metrics.SoftOPCEvaluator
  • d3rlpy.metrics.ContinuousActionDiffEvaluator
  • d3rlpy.metrics.DiscreteActionMatchEvaluator
  • d3rlpy.metrics.EnvironmentEvaluator
  • d3rlpy.metrics.CompareContinuousActionDiffEvaluator
  • d3rlpy.metrics.CompareDiscreteActionMatchEvaluator
• Off-Policy Evaluation
  • For continuous control algorithms
  • For discrete control algorithms
• Logging
  • LoggerAdapter
  • LoggerAdapterFactory
• Online Training
  • Explorers