commonpower.control.configs.algorithms.MAPPOBaseConfig

class MAPPOBaseConfig(*, algorithm_name: str, seed: int, num_env_steps: int, cuda: bool = False, cuda_deterministic: bool = True, n_training_threads: int = 1, n_rollout_threads: int = 1, n_eval_rollout_threads: int = 1, episode_length: int = 24, share_policy: bool = False, use_centralized_V: bool = True, hidden_size: int = 64, layer_N: int = 1, use_ReLU: bool = True, use_popart: bool = False, use_valuenorm: bool = True, use_feature_normalization: bool = False, use_orthogonal: bool = True, gain: float = 0.01, use_naive_recurrent_policy: bool = False, use_recurrent_policy: bool = False, recurrent_N: int = 1, data_chunk_length: int = 10, lr: float = 0.0005, critic_lr: float = 0.0005, opti_eps: float = 1e-05, weight_decay: float = 0.0, ppo_epoch: int = 15, use_clipped_value_loss: bool = True, clip_param: float = 0.2, num_mini_batch: int = 1, entropy_coef: float = 0.01, value_loss_coef: float = 1.0, use_max_grad_norm: bool = True, max_grad_norm: float = 10.0, use_gae: bool = True, gamma: float = 0.99, gae_lambda: float = 0.95, stacked_frames: int = 1, use_proper_time_limits: bool = False, use_huber_loss: bool = True, use_value_active_masks: bool = False, use_policy_active_masks: bool = False, huber_delta: float = 10.0, use_linear_lr_decay: bool = False, log_interval: int = 1, use_eval: bool = False, eval_interval: int = 25, eval_episodes: int = 32, ifi: float = 0.1, penalty_factor: float = 0.0, log_std_init: float = 0.0)

Bases: BaseModel

Create a new model by parsing and validating input data from keyword arguments.

Raises [ValidationError][pydantic_core.ValidationError] if the input data cannot be validated to form a valid model.

self is explicitly positional-only to allow self as a field name.

Methods

construct
copy	Returns a copy of the model.
dict
from_orm
json
model_construct	Creates a new instance of the Model class with validated data.
model_copy	Usage docs: https://docs.pydantic.dev/2.10/concepts/serialization/#model_copy
model_dump	Usage docs: https://docs.pydantic.dev/2.10/concepts/serialization/#modelmodel_dump
model_dump_json	Usage docs: https://docs.pydantic.dev/2.10/concepts/serialization/#modelmodel_dump_json
model_json_schema	Generates a JSON schema for a model class.
model_parametrized_name	Compute the class name for parametrizations of generic classes.
model_post_init	Override this method to perform additional initialization after __init__ and model_construct.
model_rebuild	Try to rebuild the pydantic-core schema for the model.
model_validate	Validate a pydantic model instance.
model_validate_json	Usage docs: https://docs.pydantic.dev/2.10/concepts/json/#json-parsing
model_validate_strings	Validate the given object with string data against the Pydantic model.
parse_file
parse_obj
parse_raw
schema
schema_json
update_forward_refs
validate

Attributes

model_computed_fields
model_config	Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].
model_extra	Get extra fields set during validation.
model_fields
model_fields_set	Returns the set of fields that have been explicitly set on this model instance.
algorithm_name
seed
num_env_steps
cuda
cuda_deterministic
n_training_threads
n_rollout_threads
n_eval_rollout_threads
episode_length
share_policy
use_centralized_V
hidden_size
layer_N
use_ReLU
use_popart
use_valuenorm
use_feature_normalization
use_orthogonal
gain
use_naive_recurrent_policy
use_recurrent_policy
recurrent_N
data_chunk_length
lr
critic_lr
opti_eps
weight_decay
ppo_epoch
use_clipped_value_loss
clip_param
num_mini_batch
entropy_coef
value_loss_coef
use_max_grad_norm
max_grad_norm
use_gae
gamma
gae_lambda
stacked_frames
use_proper_time_limits
use_huber_loss
use_value_active_masks
use_policy_active_masks
huber_delta
use_linear_lr_decay
log_interval
use_eval
eval_interval
eval_episodes
ifi
penalty_factor
log_std_init

model_config: ClassVar[ConfigDict] = {}

Configuration for the model, should be a dictionary conforming to [ConfigDict][pydantic.config.ConfigDict].