MCPcopy Create free account

hub / github.com/2019ChenGong/Offline_RL_Poisoner / functions

Functions1,427 in github.com/2019ChenGong/Offline_RL_Poisoner

⨍ Functions 1,427◇ Types & classes 176↳ Endpoints 37
↓ 2 callersFunctioncallback
(value)
tests/test_containers.py:30
↓ 2 callersFunctioncheck_action_scaler
Checks value and returns Scaler object. Returns: action scaler object.
d3rlpy/argument_utility.py:64
↓ 2 callersFunctioncheck_reward_scaler
Checks value and returns Scaler object. Returns: reward scaler object.
d3rlpy/argument_utility.py:80
↓ 2 callersFunctioncheck_scaler
Checks value and returns Scaler object. Returns: scaler object.
d3rlpy/argument_utility.py:48
↓ 2 callersMethodclip_episode
Clips the current episode.
d3rlpy/online/buffers.py:101
↓ 2 callersMethodcollect
Collects data via interaction with environment. If ``buffer`` is not given, ``ReplayBuffer`` will be internally created. Args:
d3rlpy/algos/base.py:275
↓ 2 callersMethodcommit
(self, epoch: int, step: int)
d3rlpy/logger.py:115
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/dynamics.py:113
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/dynamics.py:211
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/qr_q_function.py:49
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/qr_q_function.py:121
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/iqn_q_function.py:97
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/iqn_q_function.py:191
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/fqf_q_function.py:80
↓ 2 callersMethodcompute_error
( self, observations: torch.Tensor, actions: torch.Tensor, rewards: torch.Tens
d3rlpy/models/torch/q_functions/fqf_q_function.py:208
↓ 2 callersFunctioncompute_huber_loss
( y: torch.Tensor, target: torch.Tensor, beta: float = 1.0 )
d3rlpy/models/torch/q_functions/utility.py:27
↓ 2 callersFunctioncompute_max_with_n_actions
( x: torch.Tensor, actions: torch.Tensor, q_func: EnsembleContinuousQFunction, lam: float, )
d3rlpy/models/torch/q_functions/__init__.py:66
↓ 2 callersFunctioncompute_max_with_n_actions_and_indices
Returns weighted target value from sampled actions. This calculation is proposed in BCQ paper for the first time. `x` should be shaped with `(
d3rlpy/models/torch/q_functions/__init__.py:8
↓ 2 callersFunctioncompute_quantile_huber_loss
( y: torch.Tensor, target: torch.Tensor, taus: torch.Tensor )
d3rlpy/models/torch/q_functions/utility.py:35
↓ 2 callersMethodcompute_target
( self, batch: TorchMiniBatch, next_actions: torch.Tensor )
d3rlpy/ope/torch/fqe_impl.py:132
↓ 2 callersMethodcompute_target
(self, batch: TorchMiniBatch)
d3rlpy/algos/torch/dqn_impl.py:128
↓ 2 callersMethodcompute_target
( self, x: torch.Tensor, action: Optional[torch.Tensor] = None )
d3rlpy/models/torch/q_functions/qr_q_function.py:78
↓ 2 callersMethodcompute_target
( self, x: torch.Tensor, action: Optional[torch.Tensor] = None )
d3rlpy/models/torch/q_functions/iqn_q_function.py:126
↓ 2 callersMethodcompute_target
( self, x: torch.Tensor, action: Optional[torch.Tensor] = None )
d3rlpy/models/torch/q_functions/fqf_q_function.py:147
↓ 2 callersMethodcompute_target
( self, x: torch.Tensor, action: Optional[torch.Tensor] = None )
d3rlpy/models/torch/q_functions/mean_q_function.py:44
↓ 2 callersFunctioncreate_categorical_policy
( observation_shape: Sequence[int], action_size: int, encoder_factory: EncoderFactory, )
d3rlpy/models/builders.py:109
↓ 2 callersFunctioncreate_deterministic_regressor
( observation_shape: Sequence[int], action_size: int, encoder_factory: EncoderFactory, )
d3rlpy/models/builders.py:152
↓ 2 callersFunctioncreate_probabilistic_ensemble_dynamics_model
( observation_shape: Sequence[int], action_size: int, encoder_factory: EncoderFactory, n_ensem
d3rlpy/models/builders.py:181
↓ 2 callersFunctioncreate_probablistic_regressor
( observation_shape: Sequence[int], action_size: int, encoder_factory: EncoderFactory, min_log
d3rlpy/models/builders.py:161
↓ 2 callersFunctioncreate_value_function
( observation_shape: Sequence[int], encoder_factory: EncoderFactory )
d3rlpy/models/builders.py:174
↓ 2 callersMethoddist
(self, x: torch.Tensor)
d3rlpy/models/torch/imitators.py:207
↓ 2 callersFunctionevaluate_on_environment_rob_test
Returns scorer function of evaluation on environment. This function returns scorer function, which is suitable to the standard scikit-learn s
d3rlpy/metrics/scorer.py:628
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/reward_scalers.py:252
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/reward_scalers.py:347
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/reward_scalers.py:442
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/action_scalers.py:161
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/scalers.py:187
↓ 2 callersMethodfit
(self, transitions: List[Transition])
d3rlpy/preprocessing/scalers.py:320
↓ 2 callersMethodfit_with_env
Gets scaling parameters from environment. Args: env: gym environment.
d3rlpy/preprocessing/scalers.py:25
↓ 2 callersMethodfitter
Iterate over epochs steps to train with the given dataset. At each iteration algo methods and properties can be changed or queried.
d3rlpy/base.py:428
↓ 2 callersMethodforward
( self, x: torch.Tensor, deterministic: bool = False, with_log_prob: bool = Fa
d3rlpy/models/torch/policies.py:171
↓ 2 callersMethodforward
( self, x: torch.Tensor, deterministic: bool = False, with_log_prob: bool = Fa
d3rlpy/models/torch/policies.py:260
↓ 2 callersMethodforward
(self, x: torch.Tensor, action: torch.Tensor)
d3rlpy/models/torch/q_functions/mean_q_function.py:71
↓ 2 callersMethodget_action_type
(self)
d3rlpy/ope/fqe.py:173
↓ 2 callersMethodget_action_type
(self)
d3rlpy/algos/bc.py:160
↓ 2 callersFunctionget_d4rl
Returns d4rl dataset and envrironment. The dataset is provided through d4rl. .. code-block:: python from d3rlpy.datasets import get
d3rlpy/datasets.py:213
↓ 2 callersMethodget_feature_size
(self)
tests/models/torch/model_test.py:66
↓ 2 callersMethodget_logstd_parameter
(self)
d3rlpy/models/torch/policies.py:238
↓ 2 callersMethodget_next
(self)
d3rlpy/iterators/base.py:87
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/preprocessing/reward_scalers.py:276
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/preprocessing/reward_scalers.py:370
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/preprocessing/action_scalers.py:216
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/encoders.py:155
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/encoders.py:238
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/encoders.py:319
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/encoders.py:399
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/q_functions.py:117
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/q_functions.py:156
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/q_functions.py:224
↓ 2 callersMethodget_params
(self, deep: bool = False)
d3rlpy/models/q_functions.py:303
↓ 2 callersFunctionget_plt
()
d3rlpy/cli.py:32
↓ 2 callersFunctionope_tester
(ope, observation_shape, action_size=2)
tests/ope/test_fqe.py:12
↓ 2 callersFunctionparallel
()
d3rlpy/context.py:12
↓ 2 callersFunctionpick_value_by_action
( values: torch.Tensor, action: torch.Tensor, keepdim: bool = False )
d3rlpy/models/torch/q_functions/utility.py:7
↓ 2 callersMethodpredict
Returns actions. Args: observation: observation. state: this argument is just ignored. mask: this argumen
d3rlpy/wrappers/sb3.py:28
↓ 2 callersFunctionprint_stats
(path: str)
d3rlpy/cli.py:22
↓ 2 callersMethodreset
(self, **kwargs: Any)
d3rlpy/envs/wrappers.py:66
↓ 2 callersFunctionreset_optimizer_states
(impl: Any)
d3rlpy/torch_utility.py:113
↓ 2 callersMethodsample_action
(self, x: Union[np.ndarray, List[Any]])
d3rlpy/algos/random_policy.py:60
↓ 2 callersMethodsample_action
(self, x: Union[np.ndarray, List[Any]])
d3rlpy/algos/random_policy.py:121
↓ 2 callersMethodsample_action
BCQ does not support sampling action.
d3rlpy/algos/bcq.py:281
↓ 2 callersMethodsample_action
(self, x: Union[np.ndarray, List[Any]])
d3rlpy/online/iterators.py:30
↓ 2 callersMethodsample_n
( self, x: torch.Tensor, n: int, with_squash: bool = True )
d3rlpy/models/torch/imitators.py:93
↓ 2 callersMethodsample_n
(self, x: torch.Tensor, n: int)
d3rlpy/models/torch/policies.py:32
↓ 2 callersMethodsave_model
(self, fname: str)
d3rlpy/base.py:60
↓ 2 callersMethodsave_params
Saves configurations as params.json. Args: logger: logger object.
d3rlpy/base.py:810
↓ 2 callersFunctionset_eval_mode
(impl: Any)
d3rlpy/torch_utility.py:53
↓ 2 callersFunctionset_train_mode
(impl: Any)
d3rlpy/torch_utility.py:60
↓ 2 callersMethodstep
( self, action: Union[int, np.ndarray] )
d3rlpy/envs/wrappers.py:54
↓ 2 callersFunctionto_cpu
(impl: Any)
d3rlpy/torch_utility.py:74
↓ 2 callersFunctionto_cuda
(impl: Any, device: str)
d3rlpy/torch_utility.py:67
↓ 2 callersFunctiontorch_impl_tester
(impl, discrete)
tests/ope/torch/test_fqe_impl.py:17
↓ 2 callersMethodtransform
(self, reward: torch.Tensor)
d3rlpy/preprocessing/reward_scalers.py:261
↓ 2 callersMethodtransform
(self, reward: torch.Tensor)
d3rlpy/preprocessing/reward_scalers.py:356
↓ 2 callersMethodtransform
(self, x: torch.Tensor)
d3rlpy/preprocessing/scalers.py:350
↓ 2 callersFunctionunfreeze
(impl: Any)
d3rlpy/torch_utility.py:89
↓ 2 callersMethodupdate
Update parameters with mini-batch of data. Args: batch: mini-batch data. Returns: dictionary of metrics.
d3rlpy/base.py:737
↓ 2 callersMethodupdate_alpha
(self, batch: TorchMiniBatch)
d3rlpy/algos/torch/cql_impl.py:121
↓ 1 callersMethod__call__
(self, x: torch.Tensor)
d3rlpy/models/torch/imitators.py:126
↓ 1 callersMethod__call__
(self, x: torch.Tensor)
d3rlpy/models/torch/policies.py:60
↓ 1 callersMethod__call__
( self, x: torch.Tensor, reduction: str = "mean" )
d3rlpy/models/torch/q_functions/ensemble_q_function.py:148
↓ 1 callersMethod__eq__
(self, obj: Any)
d3rlpy/gpu.py:63
↓ 1 callersMethod__init__
( self, *, actor_learning_rate: float = 1e-4, critic_learning_rate: float = 3e
d3rlpy/algos/cql.py:136
↓ 1 callersMethod__init__
( self, *, distribution: str = "uniform", normal_std: float = 1.0, act
d3rlpy/algos/random_policy.py:30
↓ 1 callersMethod__init__
( self, *, actor_learning_rate: float = 3e-4, critic_learning_rate: float = 3e
d3rlpy/algos/sac.py:118
↓ 1 callersMethod__init__
( self, *, actor_learning_rate: float = 1e-3, critic_learning_rate: float = 1e
d3rlpy/algos/bcq.py:168
↓ 1 callersMethod__init__
( self, *, actor_learning_rate: float = 1e-4, critic_learning_rate: float = 1e
d3rlpy/algos/plas.py:102
↓ 1 callersMethod__init__
( self, observation_shape: Sequence[int], action_size: int, actor_learning_rat
d3rlpy/algos/torch/bcq_impl.py:44
↓ 1 callersMethod__init__
( self, observation_shape: Sequence[int], action_size: int, actor_learning_rat
d3rlpy/algos/torch/sac_impl.py:44
↓ 1 callersMethod__init__
( self, observation_shape: Sequence[int], action_size: int, actor_learning_rat
d3rlpy/algos/torch/cql_impl.py:33
← previousnext →201–300 of 1,427, ranked by callers