rlpack.utils.internal_code_setup.InternalCodeSetup Class Reference

Inheritance diagram for rlpack.utils.internal_code_setup.InternalCodeSetup:

Collaboration diagram for rlpack.utils.internal_code_setup.InternalCodeSetup:

Public Member Functions
def	__init__ (self)
None	check_validity_of_apply_norm_code (self, int apply_norm)
	Check of validity of the apply_norm code. More...
None	check_validity_of_apply_norm_to_code (self, int apply_norm_to)
	Check of validity of the apply_norm_to code. More...
int	get_apply_norm_mode_code (self, str apply_norm)
	This method retrieves the apply_norm code from the given string. More...
int	get_apply_norm_to_mode_code (self, List[str] apply_norm_to)
	This method retrieves the apply_norm code_to from the given string. More...
int	get_prioritization_code (self, str prioritization_strategy)
	This method retrieves the prioritization code for corresponding strategy passed as string in prioritized parameters. More...
def	__init__ (self)

Additional Inherited Members
Data Fields inherited from rlpack.utils.base.internal_code_register.InternalCodeRegister
	norm_mode_codes
	The mapping between given keyword and normalisation method codes. More...
	norm_to_mode_codes
	The mapping between given keyword and normalisation quantity (apply_norm_to) codes. More...
	prioritization_strategy_codes
	The mapping between prioritization strategy keywords and prioritization strategy codes. More...

Constructor & Destructor Documentation

__init__()

def rlpack.utils.internal_code_setup.InternalCodeSetup.__init__

(

self

)

Reimplemented from rlpack.utils.base.internal_code_register.InternalCodeRegister.

Member Function Documentation

check_validity_of_apply_norm_code()

None rlpack.utils.internal_code_setup.InternalCodeSetup.check_validity_of_apply_norm_code	(		self,
		int	apply_norm
	)

Check of validity of the apply_norm code.

Raises ValueError if code is invalid.

Parameters

apply_norm

int: apply_norm code to check

check_validity_of_apply_norm_to_code()

None rlpack.utils.internal_code_setup.InternalCodeSetup.check_validity_of_apply_norm_to_code	(		self,
		int	apply_norm_to
	)

Check of validity of the apply_norm_to code.

Raises ValueError if code is invalid.

Parameters

apply_norm_to

int: apply_norm_to code to check

get_apply_norm_mode_code()

int rlpack.utils.internal_code_setup.InternalCodeSetup.get_apply_norm_mode_code	(		self,
		str	apply_norm
	)

This method retrieves the apply_norm code from the given string.

This code is to be supplied to agents.

Parameters

apply_norm

str: The apply_norm string, specifying the normalization techniques to be used. *See the notes below to see the accepted values.

Returns

(int): The code corresponding to the supplied valid apply_norm.

• NOTE The value accepted for apply_norm are:
  • "none": No normalization
  • "min_max": Min-Max normalization
  • "standardize": Standardization.
  • "p_norm": P-Normalization

get_apply_norm_to_mode_code()

int rlpack.utils.internal_code_setup.InternalCodeSetup.get_apply_norm_to_mode_code	(		self,
		List[str]	apply_norm_to
	)

This method retrieves the apply_norm code_to from the given string.

This code is to be supplied to agents.

Parameters

apply_norm_to

List[str]: The apply_norm_to list, specifying the quantities on which we wish to apply normalization specified by apply_norm *See the notes below to see the accepted values.

Returns

int: The code corresponding to the supplied valid apply_norm_to.

*NOTE The value accepted for apply_norm_to are:

• ["none"]: Don't apply normalization to any quantity.
• ["states"]: Apply normalization to states.
• ["rewards"]: Apply normalization to rewards.
• ["td"]: Apply normalization for TD values.
• ["states", "rewards"]: Apply normalization to states and rewards.
• ["states", "td"]: Apply normalization to states and TD values.

get_prioritization_code()

int rlpack.utils.internal_code_setup.InternalCodeSetup.get_prioritization_code	(		self,
		str	prioritization_strategy
	)

This method retrieves the prioritization code for corresponding strategy passed as string in prioritized parameters.

Parameters

prioritization_strategy

str: A dictionary containing memory prioritization parameters for agents that may use it *See the notes below to see the accepted values.

Returns

int: The prioritization code for corresponding string value.

*NOTE: The accepted values for prioritization_strategy are as follows:

• "uniform": No prioritization is done, i.e., uniform sampling takes place; Code: 0.
• "proportional": Proportional prioritization takes place when sampling transition; Code: 1.
• "rank-based": Rank based prioritization takes place when sampling transitions; Code: 2.