mlos_bench.tunables.covariant_group module¶

Tunable parameter definition.

class mlos_bench.tunables.covariant_group.CovariantTunableGroup(name: str, config: dict)¶

Bases: object

A collection of tunable parameters.

Changing any of the parameters in the group incurs the same cost of the experiment.

Attributes:

cost: Get the cost of changing the values in the covariant group.
name: Get the name of the covariant group.

Methods

`copy`()	Deep copy of the CovariantTunableGroup object.
`equals_defaults`(other)	Checks to see if the other CovariantTunableGroup is the same, ignoring the current values of the two groups' Tunables.
`get_current_cost`()	Get the cost of the experiment given current tunable values.
`get_names`()	Get the names of all tunables in the group.
`get_tunable`(tunable)	Access the entire Tunable in a group (not just its value).
`get_tunable_values_dict`()	Get current values of all tunables in the group as a dict.
`get_tunables`()	Gets the set of tunables for this CovariantTunableGroup.
`is_defaults`()	Checks whether the currently assigned values of all tunables are at their defaults.
`is_updated`()	Check if any of the tunable values in the group has been updated.
`reset_is_updated`()	Clear the update flag.
`restore_defaults`()	Restore all tunable parameters to their default values.

copy() → CovariantTunableGroup¶

Deep copy of the CovariantTunableGroup object.

Returns:

groupCovariantTunableGroup: A new instance of the CovariantTunableGroup object that is a deep copy of the original one.

property cost: int¶

Get the cost of changing the values in the covariant group. This value is a constant. Use get_current_cost() to get the cost given the group update status.

Returns:

costint: Cost of changing the values in the covariant group.

equals_defaults(other: CovariantTunableGroup) → bool¶

Checks to see if the other CovariantTunableGroup is the same, ignoring the current values of the two groups’ Tunables.

Parameters:

otherCovariantTunableGroup: A covariant tunable group object to compare to.

Returns:

are_equalbool: True if the two CovariantTunableGroup objects’ metadata are the same, False otherwise.

get_current_cost() → int¶

Get the cost of the experiment given current tunable values.

Returns:

costint: Cost of the experiment or 0 if parameters have not been updated.

get_names() → Iterable[str]¶: Get the names of all tunables in the group.

get_tunable(tunable: str | Tunable) → Tunable¶

Access the entire Tunable in a group (not just its value). Throw KeyError if the tunable is not in the group.

Parameters:

tunablestr: Name of the tunable parameter.

Returns:

Tunable: An instance of the Tunable parameter.

get_tunable_values_dict() → Dict[str, int | float | str | None]¶

Get current values of all tunables in the group as a dict.

Returns:

tunablesDict[str, TunableValue]

get_tunables() → Iterable[Tunable]¶

Gets the set of tunables for this CovariantTunableGroup.

Returns:

Iterable[Tunable]

is_defaults() → bool¶

Checks whether the currently assigned values of all tunables are at their defaults.

Returns:

bool

is_updated() → bool¶

Check if any of the tunable values in the group has been updated.

Returns:

is_updatedbool: True if any of the tunable values in the group has been updated, False otherwise.

property name: str¶

Get the name of the covariant group.

Returns:

namestr: Name (i.e., a string id) of the covariant group.

reset_is_updated() → None¶

Clear the update flag.

That is, state that running an experiment with the current values of the tunables in this group has no extra cost.

restore_defaults() → None¶: Restore all tunable parameters to their default values.