Index

_

__contains__() (PolicyObjective class method) (RewardDomain class method) __dir__() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] __getitem__() (PolicyObjective class method) (RewardDomain class method) __init__() (AbruptShift method) (ActionValuePolicy method) (BackendCompatibilityReport method) (BanditEnvironment method) (BatchTargetPolicy method) (BenchmarkConfig method) (BenchmarkResult method) (BernoulliBayesianUCBPolicy method) (BernoulliReward method) (BernoulliThompsonSamplingPolicy method) (BetaArmPrior method) (BootstrapConfig method) (CallableContextProvider method) (ChangePointUCBPolicy method) (ClassicEnvironment method) (CompatibilityError method) (ConfidenceMethod method) (ContextualEnvironment method) (ContextualPolicy method) (CrossFittedRewardModel method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EstimateMethod method) (EstimatorConfig method) (EXP3Policy method) (Experiment method) (ExperimentConfig method) (FixedContextProvider method) (GaussianArmPrior method) (GaussianBayesianUCBPolicy method) (GaussianContextProvider method) (GaussianReward method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GradualDrift method) (GreedyPolicy method) (IntervalEstimate method) (KLUCBPolicy method) (LinearContextualEnvironment method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LoggedBanditDataset method) (LoggingScheme method) (LogisticContextualBanditPolicy method) (LogisticContextualEnvironment method) (MedianEliminationPolicy method) (MOSSPolicy method) (OfflineEstimate method) (OverlapError method) (OverlapStatus method) (PageHinkleyUCBPolicy method) (Policy method) (PolicyCapabilities method) (PolicyComparison method) (PolicySummary method) (PolicyValueEstimator method) (ProbabilityDrift method) (PyMABError method) (RandomArmSwap method) (RandomPolicy method) (ResamplingUnit method), [1] (RunProvenance method) (SequentialReplayResult method) (SerializationError method) (SimulationResult method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (StationaryDynamics method) (SuccessiveEliminationPolicy method) (TargetPolicy method) (UCBPolicy method) (UCBStats method) (UniformArmPrior method) (UniformReward method) (ValidationError method) (WeightDiagnostics method)

__iter__() (PolicyObjective class method) (RewardDomain class method) __len__() (PolicyObjective class method) (RewardDomain class method) __new__() (CompatibilityError class method) (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapError class method) (OverlapStatus method) (PyMABError class method) (ResamplingUnit method), [1] (SerializationError class method) (ValidationError class method) __new_member__() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] _generate_next_value_() (ConfidenceMethod static method) (EstimateMethod static method) (LoggingScheme static method) (OverlapStatus static method) (ResamplingUnit static method), [1] _parity_state() (ActionValuePolicy method)

A

a (LinearThompsonSamplingPolicy attribute) (LinUCBPolicy attribute) AbruptShift (class in pymab.environments) acceptance_rate (SequentialReplayResult property) acceptance_scale (SequentialReplayResult attribute) accepted_actions (SequentialReplayResult attribute) accepted_event_indices (SequentialReplayResult attribute) accepted_rewards (SequentialReplayResult attribute) action_probabilities() (EXP3Policy method) (SoftmaxPolicy method) actions (LoggedBanditDataset attribute) (SimulationResult attribute) ActionValuePolicy (class in pymab.policies.policy) active (SuccessiveEliminationPolicy attribute) add_note() (CompatibilityError method) (OverlapError method) (PyMABError method) (SerializationError method) (ValidationError method)

ADEQUATE (OverlapStatus attribute) advance() (BanditEnvironment method) (ClassicEnvironment method) alpha (BetaArmPrior attribute) apply() (AbruptShift method) (EnvironmentDynamics method) (GradualDrift method) (ProbabilityDrift method) (RandomArmSwap method) (StationaryDynamics method) args (CompatibilityError attribute) (OverlapError attribute) (PyMABError attribute) (SerializationError attribute) (ValidationError attribute) arm_means (SimulationResult attribute) ArmPrior (class in pymab.distributions) average_reward (SequentialReplayResult property) average_reward_by_step (SimulationResult property) average_reward_by_step() (in module pymab.metrics)

B

b (LinearThompsonSamplingPolicy attribute) (LinUCBPolicy attribute) backend (BernoulliBayesianUCBPolicy property) (BernoulliThompsonSamplingPolicy property) (ChangePointUCBPolicy property) (CUSUMUCBPolicy property) (DecayingEpsilonGreedyPolicy property) (DiscountedBernoulliThompsonSamplingPolicy property) (DiscountedUCBPolicy property) (EpsilonGreedyPolicy property) (EXP3Policy property) (ExperimentConfig attribute) (GaussianBayesianUCBPolicy property) (GaussianThompsonSamplingPolicy property) (GradientBanditPolicy property) (GreedyPolicy property) (KLUCBPolicy property) (LinearEpsilonGreedyPolicy property) (LinearThompsonSamplingPolicy property) (LinUCBPolicy property) (LogisticContextualBanditPolicy property) (MedianEliminationPolicy property) (MOSSPolicy property) (PageHinkleyUCBPolicy property) (RandomPolicy property) (RunProvenance attribute) (SlidingWindowBernoulliThompsonSamplingPolicy property) (SlidingWindowUCBPolicy property) (SoftmaxPolicy property) (SuccessiveEliminationPolicy property) (UCBPolicy property)

backend_compatibility() (Experiment method) BackendCompatibilityReport (class in pymab.simulation) BanditEnvironment (class in pymab.environments) baseline (BenchmarkConfig attribute) baseline_id (PolicyComparison attribute) BatchTargetPolicy (class in pymab.offline) BenchmarkConfig (class in pymab.benchmarking) BenchmarkResult (class in pymab.benchmarking) BernoulliBayesianUCBPolicy (class in pymab.policies.bayesian_ucb) BernoulliReward (class in pymab.distributions) BernoulliThompsonSamplingPolicy (class in pymab.policies.thompson_sampling) best_arm (MedianEliminationPolicy property) BEST_ARM (PolicyObjective attribute) best_arm (SuccessiveEliminationPolicy property) best_arm_identification_rate_by_step() (in module pymab.metrics) beta (BetaArmPrior attribute) BetaArmPrior (class in pymab.distributions) BINARY (RewardDomain attribute) bonuses (UCBStats attribute) bootstrap (BenchmarkConfig attribute) (EstimatorConfig attribute) bootstrap_mean_interval() (in module pymab.statistics) BootstrapConfig (class in pymab.statistics)

C

CallableContextProvider (class in pymab.environments) capabilities (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (ContextualPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) (GradientBanditPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (LinearEpsilonGreedyPolicy attribute) (LinearThompsonSamplingPolicy attribute) (LinUCBPolicy attribute) (LogisticContextualBanditPolicy attribute) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (Policy attribute) (RandomPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute) capitalize() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] capture() (RunProvenance class method) casefold() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] center() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] change_counts (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) ChangePointUCBPolicy (class in pymab.policies.change_detection) choose_argmax() (in module pymab.policies.policy) ci_lower (IntervalEstimate attribute) (OfflineEstimate attribute) ci_upper (IntervalEstimate attribute) (OfflineEstimate attribute) ClassicEnvironment (class in pymab.environments) clipped_fraction (OfflineEstimate property) (WeightDiagnostics attribute) clipping_threshold (WeightDiagnostics attribute) clone() (AbruptShift method) (ActionValuePolicy method) (BanditEnvironment method) (BernoulliBayesianUCBPolicy method) (BernoulliReward method) (BernoulliThompsonSamplingPolicy method) (CallableContextProvider method) (ChangePointUCBPolicy method) (ClassicEnvironment method) (ContextProvider method) (ContextualEnvironment method) (ContextualPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EnvironmentDynamics method) (EpsilonGreedyPolicy method) (EXP3Policy method) (FixedContextProvider method) (GaussianBayesianUCBPolicy method) (GaussianContextProvider method) (GaussianReward method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GradualDrift method) (GreedyPolicy method) (KLUCBPolicy method) (LinearContextualEnvironment method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (LogisticContextualEnvironment method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (Policy method) (ProbabilityDrift method) (RandomArmSwap method) (RandomPolicy method) (RewardModel method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (StationaryDynamics method) (SuccessiveEliminationPolicy method) (UCBPolicy method) (UniformReward method)

CLUSTER (ResamplingUnit attribute), [1] clusters (LoggedBanditDataset attribute) compare() (in module pymab.benchmarking) compare_to_baseline() (BenchmarkResult method) CompatibilityError compatible (BackendCompatibilityReport property) component_snapshot() (in module pymab.provenance) confidence_level (BootstrapConfig attribute) (IntervalEstimate attribute) (OfflineEstimate attribute) confidence_method (IntervalEstimate attribute) (OfflineEstimate attribute) ConfidenceMethod (class in pymab.statistics) config (BenchmarkResult attribute) (SimulationResult attribute) configuration (BernoulliBayesianUCBPolicy property) (BernoulliThompsonSamplingPolicy property) (ChangePointUCBPolicy property) (CUSUMUCBPolicy property) (DecayingEpsilonGreedyPolicy property) (DiscountedBernoulliThompsonSamplingPolicy property) (DiscountedUCBPolicy property) (EpsilonGreedyPolicy property) (EXP3Policy property) (GaussianBayesianUCBPolicy property) (GaussianThompsonSamplingPolicy property) (GradientBanditPolicy property) (GreedyPolicy property) (KLUCBPolicy property) (LinearEpsilonGreedyPolicy property) (LinearThompsonSamplingPolicy property) (LinUCBPolicy property) (LogisticContextualBanditPolicy property) (MedianEliminationPolicy property) (MOSSPolicy property) (PageHinkleyUCBPolicy property) (RandomPolicy property) (SlidingWindowBernoulliThompsonSamplingPolicy property) (SlidingWindowUCBPolicy property) (SoftmaxPolicy property) (SuccessiveEliminationPolicy property) (UCBPolicy property) context() (ContextualEnvironment method) (LinearContextualEnvironment method) (LogisticContextualEnvironment method) context_at() (LoggedBanditDataset method) context_digest (SimulationResult attribute) context_provider (LinearContextualEnvironment attribute) (LogisticContextualEnvironment attribute) ContextProvider (class in pymab.environments) contexts (LoggedBanditDataset attribute) (SimulationResult attribute) contextual (BanditEnvironment property) (ClassicEnvironment property) (ContextualEnvironment property) (LinearContextualEnvironment property) (LogisticContextualEnvironment property) (PolicyCapabilities attribute) ContextualEnvironment (class in pymab.environments) ContextualPolicy (class in pymab.policies.policy) count() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] counts (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (RandomPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute) (UCBStats attribute) CrossFittedRewardModel (class in pymab.offline) cumulative_regret (PolicyComparison attribute) (PolicySummary attribute) (SimulationResult property) cumulative_regret_by_replicate (SimulationResult property) cumulative_regret_by_step() (in module pymab.metrics) CUMULATIVE_REWARD (PolicyObjective attribute) cumulative_reward_by_step (SimulationResult property) cumulative_reward_by_step() (in module pymab.metrics) CUSUMUCBPolicy (class in pymab.policies.change_detection)

D

DecayingEpsilonGreedyPolicy (class in pymab.policies.epsilon_greedy) detector_counts (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) detector_means (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) discounted_counts (DiscountedUCBPolicy attribute) discounted_sums (DiscountedUCBPolicy attribute)

DiscountedBernoulliThompsonSamplingPolicy (class in pymab.policies.nonstationary) DiscountedUCBPolicy (class in pymab.policies.nonstationary) domain (BernoulliReward attribute) (GaussianReward attribute) (RewardModel attribute) (UniformReward attribute) DOUBLY_ROBUST (EstimateMethod attribute) dynamics (BanditEnvironment attribute)

E

effective_sample_size (OfflineEstimate property) (WeightDiagnostics attribute) encode() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] endswith() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] environment (RunProvenance attribute) EnvironmentDynamics (class in pymab.environments) epsilon (DecayingEpsilonGreedyPolicy property) (ProbabilityDrift attribute) EpsilonGreedyPolicy (class in pymab.policies.epsilon_greedy) equals() (LoggedBanditDataset method) (RunProvenance method) (SequentialReplayResult method) (SimulationResult method) estimate (IntervalEstimate attribute) (OfflineEstimate attribute) estimate() (PolicyValueEstimator method) estimate_policy_value() (in module pymab.offline) estimated_state_bytes() (BernoulliBayesianUCBPolicy method) (BernoulliThompsonSamplingPolicy method) (ChangePointUCBPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EXP3Policy method) (GaussianBayesianUCBPolicy method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GreedyPolicy method) (KLUCBPolicy method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (RandomPolicy method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (SuccessiveEliminationPolicy method) (UCBPolicy method)

EstimateMethod (class in pymab.offline) estimates (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (RandomPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute) (UCBStats attribute) EstimatorConfig (class in pymab.offline) EVENT (ResamplingUnit attribute), [1] EXP3Policy (class in pymab.policies.adversarial) expandtabs() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] expected_regret_by_step() (in module pymab.metrics) expected_rewards (SimulationResult attribute) expected_rewards() (BanditEnvironment method) (ClassicEnvironment method) (ContextualEnvironment method) (LinearContextualEnvironment method) (LogisticContextualEnvironment method) Experiment (class in pymab.simulation) ExperimentConfig (class in pymab.simulation)

F

failures (DiscountedBernoulliThompsonSamplingPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) final_simple_regret (PolicyComparison attribute) (PolicySummary attribute) find() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] FixedContextProvider (class in pymab.environments) format() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

format_map() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] freeze_json() (in module pymab.provenance) frequency (AbruptShift attribute) from_dict() (SimulationResult class method) from_payload() (ResultSerializer class method) from_prior() (BanditEnvironment class method) function (CallableContextProvider attribute)

G

GaussianArmPrior (class in pymab.distributions) GaussianBayesianUCBPolicy (class in pymab.policies.bayesian_ucb) GaussianContextProvider (class in pymab.environments) GaussianReward (class in pymab.distributions)

GaussianThompsonSamplingPolicy (class in pymab.policies.thompson_sampling) GradientBanditPolicy (class in pymab.policies.gradient) GradualDrift (class in pymab.environments) GreedyPolicy (class in pymab.policies.greedy)

H

half_width (UniformReward attribute) high (UniformArmPrior attribute)

horizon (ExperimentConfig attribute) (SimulationResult property)

I

index() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] indices() (KLUCBPolicy method) initial_value (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (RandomPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute) IntervalEstimate (class in pymab.statistics) IPS (EstimateMethod attribute) isalnum() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isalpha() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isascii() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

isdecimal() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isdigit() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isidentifier() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] islower() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isnumeric() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isprintable() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isspace() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] issues (BackendCompatibilityReport attribute) istitle() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] isupper() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

J

join() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

K

KLUCBPolicy (class in pymab.policies.ucb)

L

library_version (SimulationResult attribute) LinearContextualEnvironment (class in pymab.environments) LinearEpsilonGreedyPolicy (class in pymab.policies.contextual_bandits) LinearThompsonSamplingPolicy (class in pymab.policies.contextual_bandits) LinUCBPolicy (class in pymab.policies.contextual_bandits) ljust() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] load_json() (ResultSerializer class method) (SimulationResult class method) load_npz() (ResultSerializer class method) (SimulationResult class method) LoggedBanditDataset (class in pymab.offline) logging_scheme (SequentialReplayResult attribute)

LoggingScheme (class in pymab.offline) LogisticContextualBanditPolicy (class in pymab.policies.contextual_bandits) LogisticContextualEnvironment (class in pymab.environments) logit_std (ProbabilityDrift attribute) low (UniformArmPrior attribute) lower() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] lowest_mean_regret_policy (BenchmarkResult property) lstrip() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

M

maketrans() (ConfidenceMethod static method) (EstimateMethod static method) (LoggingScheme static method) (OverlapStatus static method) (ResamplingUnit static method), [1] max_chunk_elements (BootstrapConfig attribute) max_weight (OfflineEstimate property) (WeightDiagnostics attribute) mean (GaussianArmPrior attribute) (GaussianContextProvider attribute) mean_weight (OfflineEstimate property) (WeightDiagnostics attribute) means (BanditEnvironment attribute) (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) MedianEliminationPolicy (class in pymab.policies.pure_exploration) message() (BackendCompatibilityReport method) metadata (SimulationResult attribute) method (EstimatorConfig attribute) (OfflineEstimate attribute) MODEL_ONLY (OverlapStatus attribute) module pymab.benchmarking pymab.distributions pymab.environments pymab.errors pymab.metrics pymab.offline pymab.persistence pymab.plotting pymab.policies.adversarial pymab.policies.bayesian_ucb pymab.policies.change_detection pymab.policies.contextual_bandits pymab.policies.epsilon_greedy pymab.policies.gradient pymab.policies.greedy pymab.policies.nonstationary pymab.policies.policy pymab.policies.pure_exploration pymab.policies.random pymab.policies.softmax_selection pymab.policies.thompson_sampling pymab.policies.ucb pymab.provenance pymab.results pymab.simulation pymab.statistics pymab.types

MOSSPolicy (class in pymab.policies.ucb) moving_average() (in module pymab.metrics)

N

n_accepted (SequentialReplayResult property) n_arms (ActionValuePolicy attribute) (BanditEnvironment property) (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (ClassicEnvironment property) (ContextualEnvironment property) (ContextualPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianContextProvider attribute) (GaussianThompsonSamplingPolicy attribute) (GradientBanditPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (LinearContextualEnvironment property) (LinearEpsilonGreedyPolicy attribute) (LinearThompsonSamplingPolicy attribute) (LinUCBPolicy attribute) (LoggedBanditDataset attribute) (LogisticContextualBanditPolicy attribute) (LogisticContextualEnvironment property) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (Policy attribute) (RandomPolicy attribute) (SimulationResult property) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute) n_events (LoggedBanditDataset property) (OfflineEstimate attribute) (SequentialReplayResult property) n_features (BernoulliBayesianUCBPolicy attribute) (BernoulliThompsonSamplingPolicy attribute) (ChangePointUCBPolicy attribute) (ContextualEnvironment property) (ContextualPolicy attribute) (CUSUMUCBPolicy attribute) (DecayingEpsilonGreedyPolicy attribute) (DiscountedBernoulliThompsonSamplingPolicy attribute) (DiscountedUCBPolicy attribute) (EpsilonGreedyPolicy attribute) (EXP3Policy attribute) (GaussianBayesianUCBPolicy attribute) (GaussianContextProvider attribute) (GaussianThompsonSamplingPolicy attribute) (GradientBanditPolicy attribute) (GreedyPolicy attribute) (KLUCBPolicy attribute) (LinearContextualEnvironment property) (LinearEpsilonGreedyPolicy attribute) (LinearThompsonSamplingPolicy attribute) (LinUCBPolicy attribute) (LogisticContextualBanditPolicy attribute) (LogisticContextualEnvironment property) (MedianEliminationPolicy attribute) (MOSSPolicy attribute) (PageHinkleyUCBPolicy attribute) (RandomPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) (SlidingWindowUCBPolicy attribute) (SoftmaxPolicy attribute) (SuccessiveEliminationPolicy attribute) (UCBPolicy attribute)

n_observations (IntervalEstimate attribute) n_policies (SimulationResult property) n_replicates (ExperimentConfig attribute) (PolicyComparison property) (PolicySummary property) (SimulationResult property) n_resamples (BootstrapConfig attribute) negative_cusum (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) NONE (OverlapStatus attribute) NONUNIFORM (LoggingScheme attribute) numpy_version (RunProvenance attribute)

O

objective (PolicyCapabilities attribute) OfflineEstimate (class in pymab.offline) optimal_action_indicator (SimulationResult property) optimal_action_rate (PolicyComparison attribute) (PolicySummary attribute) optimal_action_rate_by_step (SimulationResult property)

optimal_action_rate_by_step() (in module pymab.metrics) optimal_mask (SimulationResult attribute) optimal_values (SimulationResult property) overlap_status (OfflineEstimate attribute) OverlapError OverlapStatus (class in pymab.offline)

P

PageHinkleyUCBPolicy (class in pymab.policies.change_detection) partition() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] PERCENTILE_BOOTSTRAP (ConfidenceMethod attribute) ph_cumulative (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) ph_minimum (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) plot_average_reward() (BenchmarkResult method) (in module pymab.plotting) plot_cumulative_regret() (BenchmarkResult method) (in module pymab.plotting) plot_optimal_action_rate() (BenchmarkResult method) (in module pymab.plotting) policies (RunProvenance attribute) Policy (class in pymab.policies.policy) policy_id (PolicyComparison attribute) (PolicySummary attribute) policy_ids (BenchmarkResult property) (SimulationResult attribute) policy_index (PolicySummary attribute) PolicyCapabilities (class in pymab.types) PolicyComparison (class in pymab.benchmarking) PolicyObjective (class in pymab.types) PolicySummary (class in pymab.benchmarking) PolicyValueEstimator (class in pymab.offline) positive_cusum (ChangePointUCBPolicy attribute) (PageHinkleyUCBPolicy attribute) precisions (GaussianBayesianUCBPolicy attribute) (GaussianThompsonSamplingPolicy attribute) predict_event() (CrossFittedRewardModel method) predicted_probabilities() (LogisticContextualBanditPolicy method) probabilities() (TargetPolicy method) probabilities_batch() (BatchTargetPolicy method) probability (RandomArmSwap attribute) ProbabilityDrift (class in pymab.environments) propensities (LoggedBanditDataset attribute) provenance (SimulationResult attribute) pymab.benchmarking module pymab.distributions module pymab.environments module pymab.errors module

pymab.metrics module pymab.offline module pymab.persistence module pymab.plotting module pymab.policies.adversarial module pymab.policies.bayesian_ucb module pymab.policies.change_detection module pymab.policies.contextual_bandits module pymab.policies.epsilon_greedy module pymab.policies.gradient module pymab.policies.greedy module pymab.policies.nonstationary module pymab.policies.policy module pymab.policies.pure_exploration module pymab.policies.random module pymab.policies.softmax_selection module pymab.policies.thompson_sampling module pymab.policies.ucb module pymab.provenance module pymab.results module pymab.simulation module pymab.statistics module pymab.types module pymab_version (RunProvenance attribute) PyMABError python_version (RunProvenance attribute)

R

RandomArmSwap (class in pymab.environments) RandomPolicy (class in pymab.policies.random) raw_effective_sample_size (WeightDiagnostics attribute) raw_max_weight (WeightDiagnostics attribute) raw_mean_weight (WeightDiagnostics attribute) REAL (RewardDomain attribute) recommend_action() (ActionValuePolicy method) (BernoulliBayesianUCBPolicy method) (BernoulliThompsonSamplingPolicy method) (ChangePointUCBPolicy method) (ContextualPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EXP3Policy method) (GaussianBayesianUCBPolicy method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GreedyPolicy method) (KLUCBPolicy method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (Policy method) (RandomPolicy method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (SuccessiveEliminationPolicy method) (UCBPolicy method) recommendation_is_optimal (SimulationResult property) recommendations (SimulationResult attribute) record_contexts (ExperimentConfig attribute) regret (SimulationResult property) removeprefix() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] removesuffix() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] replace() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] REPLICATE (ResamplingUnit attribute), [1] replicate_seeds (SimulationResult attribute) resampling_unit (IntervalEstimate attribute) (OfflineEstimate attribute) ResamplingUnit (class in pymab.offline) (class in pymab.statistics) reset() (ActionValuePolicy method) (BernoulliBayesianUCBPolicy method) (BernoulliThompsonSamplingPolicy method) (ChangePointUCBPolicy method) (ContextualPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EXP3Policy method) (GaussianBayesianUCBPolicy method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GreedyPolicy method) (KLUCBPolicy method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (Policy method) (RandomPolicy method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (SuccessiveEliminationPolicy method) (UCBPolicy method)

resolve_reward_model() (in module pymab.distributions) result (BenchmarkResult attribute) ResultSerializer (class in pymab.persistence) reward_coupling (ExperimentConfig attribute) reward_domain (BanditEnvironment property) (ClassicEnvironment property) (ContextualEnvironment property) (LinearContextualEnvironment property) (LogisticContextualEnvironment property) reward_domains (PolicyCapabilities attribute) reward_model (BanditEnvironment attribute) (ClassicEnvironment property) (ContextualEnvironment property) (LinearContextualEnvironment attribute) (LogisticContextualEnvironment attribute) RewardDomain (class in pymab.types) RewardModel (class in pymab.distributions) rewards (LoggedBanditDataset attribute) (SimulationResult attribute) rfind() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] rindex() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] rjust() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] rng_scheme (RunProvenance attribute) rpartition() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] rsplit() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] rstrip() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] run() (Experiment method) RunProvenance (class in pymab.provenance)

S

sample() (ArmPrior method) (BernoulliReward method) (BetaArmPrior method) (CallableContextProvider method) (ContextProvider method) (FixedContextProvider method) (GaussianArmPrior method) (GaussianContextProvider method) (GaussianReward method) (RewardModel method) (UniformArmPrior method) (UniformReward method) sample_one() (BernoulliReward method) (GaussianReward method) (RewardModel method) (UniformReward method) sample_rewards() (BanditEnvironment method) (LinearContextualEnvironment method) (LogisticContextualEnvironment method) save_json() (ResultSerializer class method) (SimulationResult method) save_npz() (ResultSerializer class method) (SimulationResult method) schema_version (SimulationResult attribute) scores() (LinearEpsilonGreedyPolicy method) seed (BootstrapConfig attribute) (ExperimentConfig attribute) select_action() (ActionValuePolicy method) (BernoulliBayesianUCBPolicy method) (BernoulliThompsonSamplingPolicy method) (ChangePointUCBPolicy method) (ContextualPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EXP3Policy method) (GaussianBayesianUCBPolicy method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GreedyPolicy method) (KLUCBPolicy method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (Policy method) (RandomPolicy method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (SuccessiveEliminationPolicy method) (UCBPolicy method) selected_actions (SequentialReplayResult attribute) sequential_replay() (in module pymab.offline) SequentialReplayResult (class in pymab.offline) SerializationError shift_at_step_zero (AbruptShift attribute) simple_regret (SimulationResult property) simple_regret_by_step() (in module pymab.metrics) SimulationResult (class in pymab.results) SlidingWindowBernoulliThompsonSamplingPolicy (class in pymab.policies.nonstationary) SlidingWindowUCBPolicy (class in pymab.policies.nonstationary) SNIPS (EstimateMethod attribute) softmax() (in module pymab.policies.policy) SoftmaxPolicy (class in pymab.policies.softmax_selection) split() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

splitlines() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] standard_error (IntervalEstimate attribute) (OfflineEstimate attribute) standard_error() (in module pymab.statistics) startswith() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] StationaryDynamics (class in pymab.environments) std (AbruptShift attribute) (GaussianArmPrior attribute) (GaussianContextProvider attribute) (GaussianReward attribute) (GradualDrift attribute) step (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy property) (BernoulliThompsonSamplingPolicy property) (ChangePointUCBPolicy property) (CUSUMUCBPolicy property) (DecayingEpsilonGreedyPolicy property) (DiscountedBernoulliThompsonSamplingPolicy property) (DiscountedUCBPolicy property) (EpsilonGreedyPolicy property) (EXP3Policy property) (GaussianBayesianUCBPolicy property) (GaussianThompsonSamplingPolicy property) (GradientBanditPolicy property) (GreedyPolicy property) (KLUCBPolicy property) (LinearEpsilonGreedyPolicy property) (LinearThompsonSamplingPolicy property) (LinUCBPolicy property) (LogisticContextualBanditPolicy property) (MedianEliminationPolicy property) (MOSSPolicy property) (PageHinkleyUCBPolicy property) (RandomPolicy property) (SlidingWindowBernoulliThompsonSamplingPolicy property) (SlidingWindowUCBPolicy property) (SoftmaxPolicy property) (SuccessiveEliminationPolicy property) (UCBPolicy property) strip() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] successes (DiscountedBernoulliThompsonSamplingPolicy attribute) (SlidingWindowBernoulliThompsonSamplingPolicy attribute) SuccessiveEliminationPolicy (class in pymab.policies.pure_exploration) summary() (BenchmarkResult method) supported_domains (AbruptShift attribute) (EnvironmentDynamics attribute) (GradualDrift attribute) (ProbabilityDrift attribute) (RandomArmSwap attribute) (StationaryDynamics attribute) supports() (AbruptShift method) (EnvironmentDynamics method) (GradualDrift method) (ProbabilityDrift method) (RandomArmSwap method) (StationaryDynamics method) swapcase() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

T

TargetPolicy (class in pymab.offline) thaw_json() (in module pymab.provenance) theta (LinearContextualEnvironment attribute) (LogisticContextualBanditPolicy attribute) (LogisticContextualEnvironment attribute) title() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] to_dict() (BenchmarkConfig method) (BenchmarkResult method) (BootstrapConfig method) (IntervalEstimate method) (PolicyComparison method) (PolicySummary method) (RunProvenance method) (SimulationResult method) to_pandas() (BenchmarkResult method) (SimulationResult method) total_reward (ActionValuePolicy attribute) (BernoulliBayesianUCBPolicy property) (BernoulliThompsonSamplingPolicy property) (ChangePointUCBPolicy property) (CUSUMUCBPolicy property) (DecayingEpsilonGreedyPolicy property) (DiscountedBernoulliThompsonSamplingPolicy property) (DiscountedUCBPolicy property) (EpsilonGreedyPolicy property) (EXP3Policy property) (GaussianBayesianUCBPolicy property) (GaussianThompsonSamplingPolicy property) (GradientBanditPolicy property) (GreedyPolicy property) (KLUCBPolicy property) (LinearEpsilonGreedyPolicy property) (LinearThompsonSamplingPolicy property) (LinUCBPolicy property) (LogisticContextualBanditPolicy property) (MedianEliminationPolicy property) (MOSSPolicy property) (PageHinkleyUCBPolicy property) (PolicyComparison attribute) (PolicySummary attribute) (RandomPolicy property) (SlidingWindowBernoulliThompsonSamplingPolicy property) (SlidingWindowUCBPolicy property) (SoftmaxPolicy property) (SuccessiveEliminationPolicy property) (UCBPolicy property)

translate() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]

U

UCBPolicy (class in pymab.policies.ucb) UCBStats (class in pymab.policies.ucb) UNIFORM (LoggingScheme attribute) UniformArmPrior (class in pymab.distributions) UniformReward (class in pymab.distributions) UNIT_INTERVAL (RewardDomain attribute) unknown() (RunProvenance class method) update() (ActionValuePolicy method) (BernoulliBayesianUCBPolicy method) (BernoulliThompsonSamplingPolicy method) (ChangePointUCBPolicy method) (ContextualPolicy method) (CUSUMUCBPolicy method) (DecayingEpsilonGreedyPolicy method) (DiscountedBernoulliThompsonSamplingPolicy method) (DiscountedUCBPolicy method) (EpsilonGreedyPolicy method) (EXP3Policy method) (GaussianBayesianUCBPolicy method) (GaussianThompsonSamplingPolicy method) (GradientBanditPolicy method) (GreedyPolicy method) (KLUCBPolicy method) (LinearEpsilonGreedyPolicy method) (LinearThompsonSamplingPolicy method) (LinUCBPolicy method) (LogisticContextualBanditPolicy method) (MedianEliminationPolicy method) (MOSSPolicy method) (PageHinkleyUCBPolicy method) (Policy method) (RandomPolicy method) (SlidingWindowBernoulliThompsonSamplingPolicy method) (SlidingWindowUCBPolicy method) (SoftmaxPolicy method) (SuccessiveEliminationPolicy method) (UCBPolicy method)

upper() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1] upper_confidence_bounds() (LinUCBPolicy method)

V

validate_means() (BernoulliReward method) (GaussianReward method) (RewardModel method) (UniformReward method)

validate_positive() (in module pymab.policies.policy) validate_positive_integer() (in module pymab.policies.policy) validate_probability() (in module pymab.policies.policy) ValidationError value (FixedContextProvider attribute)

W

WEAK (OverlapStatus attribute) weight_clip (EstimatorConfig attribute) WeightDiagnostics (class in pymab.offline) weights (EXP3Policy property) (OfflineEstimate attribute)

with_traceback() (CompatibilityError method) (OverlapError method) (PyMABError method) (SerializationError method) (ValidationError method)

Z

zfill() (ConfidenceMethod method) (EstimateMethod method) (LoggingScheme method) (OverlapStatus method) (ResamplingUnit method), [1]