SubstanceParameter

class baybe.parameters.substance.SubstanceParameter

Bases: DiscreteParameter

Generic substances that are treated with cheminformatics descriptors.

Only a decorrelated subset of descriptors should be used as otherwise this can result in a large number of features. For a handful of molecules, keeping only descriptors that have a maximum correlation of 0.7 reduces the number of descriptors to about 5-20. The number might be substantially higher with more labels given.

Public methods

__init__(name, data[, decorrelate, encoding])	Method generated by attrs for class SubstanceParameter.
from_dict(dictionary)	Create an object from its dictionary representation.
from_json(string)	Create an object from its JSON representation.
is_in_range(item)	Return whether an item is within the parameter range.
summary()	Return a custom summarization of the parameter.
to_dict()	Create an object's dictionary representation.
to_json()	Create an object's JSON representation.
to_searchspace()	Create a one-dimensional search space from the parameter.
to_subspace()	Create a one-dimensional search space from the parameter.
transform(series, /)	Transform parameter values from experimental to computational representation.

Public attributes and properties

comp_df	Return the computational representation of the parameter.
comp_rep_columns	The columns spanning the computational representation.
is_continuous	Boolean indicating if this is a continuous parameter.
is_discrete	Boolean indicating if this is a discrete parameter.
is_numerical	Class variable encoding whether this parameter is numeric.
values	Returns the labels of the given set of molecules.
data	A mapping that provides the SMILES strings for all available parameter values.
decorrelate	Specifies the used decorrelation mode for the parameter encoding.
encoding	An optional encoding for the parameter.
name	The name of the parameter

__init__(name: str, data, decorrelate: bool | float = True, encoding=SubstanceEncoding.MORDRED)

Method generated by attrs for class SubstanceParameter.

For details on the parameters, see Public attributes and properties.

classmethod from_dict(dictionary: dict)

Create an object from its dictionary representation.

Parameters:

dictionary (dict) – The dictionary representation.

Return type:

TypeVar(_T)

Returns:

The reconstructed object.

classmethod from_json(string: str)

Create an object from its JSON representation.

Parameters:

string (str) – The JSON representation of the object.

Return type:

TypeVar(_T)

Returns:

The reconstructed object.

is_in_range(item: Any)

Return whether an item is within the parameter range.

Parameters:

item (Any) – The item to be checked.

Return type:

bool

Returns:

True if the item is within the parameter range, False otherwise.

summary()

Return a custom summarization of the parameter.

Return type:

dict

to_dict()

Create an object’s dictionary representation.

Return type:

dict

to_json()

Create an object’s JSON representation.

Return type:

str

Returns:

The JSON representation as a string.

to_searchspace()

Create a one-dimensional search space from the parameter.

Return type:

SearchSpace

to_subspace()

Create a one-dimensional search space from the parameter.

Return type:

SubspaceDiscrete

transform(series: Series, /)

Transform parameter values from experimental to computational representation.

Parameters:

series (Series) – The parameter values to be transformed.

Return type:

DataFrame

Returns:

The transformed parameter values.

property comp_df: DataFrame

Return the computational representation of the parameter.

property comp_rep_columns: tuple[str, ...]

The columns spanning the computational representation.

data: dict[str, str]

A mapping that provides the SMILES strings for all available parameter values.

decorrelate: bool | float

Specifies the used decorrelation mode for the parameter encoding.

• False: The encoding is used as is.

• True: The encoding is decorrelated using a default correlation threshold.

• float in (0, 1): The encoding is decorrelated using the specified threshold.

encoding: SubstanceEncoding

An optional encoding for the parameter.

property is_continuous: bool

Boolean indicating if this is a continuous parameter.

property is_discrete: bool

Boolean indicating if this is a discrete parameter.

is_numerical: ClassVar[bool] = False

Class variable encoding whether this parameter is numeric.

name: str

The name of the parameter

property values: tuple

Returns the labels of the given set of molecules.