Base inventory env

Base environment with some basic funcitons

BaseInventoryEnv

 BaseInventoryEnv (mdp_info:ddopai.utils.MDPInfo,
                   postprocessors:list[object]|None=None,
                   mode:str='train', return_truncation:str=True,
                   dataloader:ddopai.dataloaders.base.BaseDataLoader=None,
                   horizon_train:int=100, underage_cost:Union[numpy.ndarra
                   y,ddopai.utils.Parameter,int,float]=1, overage_cost:Uni
                   on[numpy.ndarray,ddopai.utils.Parameter,int,float]=0)

Base class for inventory management environments. This class inherits from BaseEnvironment.

	Type	Default	Details
mdp_info	MDPInfo
postprocessors	list[object] \| None	None	default is empty list
mode	str	train	Initial mode (train, val, test) of the environment
return_truncation	str	True	whether to return a truncated condition in step function
dataloader	BaseDataLoader	None	dataloader for the environment
horizon_train	int	100	horizon for training mode
underage_cost	Union	1	underage cost per unit
overage_cost	Union	0	overage cost per unit (zero in most cases)
Returns	None

source

BaseInventoryEnv.set_observation_space

 BaseInventoryEnv.set_observation_space (shape:tuple,
                                         low:Union[numpy.ndarray,float]=-
                                         inf, high:Union[numpy.ndarray,flo
                                         at]=inf,
                                         samples_dim_included=True)

Set the observation space of the environment. This is a standard function for simple observation spaces. For more complex observation spaces, this function should be overwritten. Note that it is assumped that the first dimension is n_samples that is not relevant for the observation space.

	Type	Default	Details
shape	tuple		shape of the dataloader features
low	Union	-inf	lower bound of the observation space
high	Union	inf	upper bound of the observation space
samples_dim_included	bool	True	whether the first dimension of the shape input is the number of samples
Returns	None

source

BaseInventoryEnv.set_action_space

 BaseInventoryEnv.set_action_space (shape:tuple,
                                    low:Union[numpy.ndarray,float]=-inf,
                                    high:Union[numpy.ndarray,float]=inf,
                                    samples_dim_included=True)

Set the action space of the environment. This is a standard function for simple action spaces. For more complex action spaces, this function should be overwritten. Note that it is assumped that the first dimension is n_samples that is not relevant for the action space.

	Type	Default	Details
shape	tuple		shape of the dataloader target
low	Union	-inf	lower bound of the observation space
high	Union	inf	upper bound of the observation space
samples_dim_included	bool	True	whether the first dimension of the shape input is the number of samples
Returns	None

source

BaseInventoryEnv.reset

 BaseInventoryEnv.reset (start_index:int|str=None,
                         state:numpy.ndarray=None)

Reset function for the Newsvendor problem. It will return the first observation and demand. For val and test modes, it will by default reset to 0, while for the train mode it depends on the paramter “horizon_train” whether a random point in the training data is selected or 0

	Type	Default	Details
start_index	int \| str	None	index to start from
state	ndarray	None	initial state
Returns	Tuple

source

BaseInventoryEnv.get_observation

 BaseInventoryEnv.get_observation ()

Return the current observation. This function is for the simple case where the observation is only an x,y pair. For more complex observations, this function should be overwritten.