Vartype

Enumeration of valid variable types for binary quadratic models.

Examples

Vartype is an Enum. Each vartype has a value and a name.

>>> vartype = dimod.SPIN
>>> vartype.name
'SPIN'
>>> vartype.value == {-1, +1}
True

>>> vartype = dimod.BINARY
>>> vartype.name
'BINARY'
>>> vartype.value == {0, 1}
True

The as_vartype() function allows the user to provide several convenient forms.

>>> from dimod import as_vartype

>>> as_vartype(dimod.SPIN) is dimod.SPIN
True
>>> as_vartype('SPIN') is dimod.SPIN
True
>>> as_vartype({-1, 1}) is dimod.SPIN
True

>>> as_vartype(dimod.BINARY) is dimod.BINARY
True
>>> as_vartype('BINARY') is dimod.BINARY
True
>>> as_vartype({0, 1}) is dimod.BINARY
True

class Vartype(value)

An Enum over the types of variables for quadratic models.

SPIN

Vartype for spin-valued binary quadratic models and variables of quadratic models that have values that are either -1 or 1.

BINARY

Vartype for binary quadratic models and variables of quadratic models that have values that are either 0 or 1.

INTEGER

Vartype for variables in quadratic models that have values of type int.

REAL

Vartype for variables in quadratic models that have values of type float.

as_vartype(vartype: Literal[Vartype.SPIN, 'SPIN', Vartype.BINARY, 'BINARY', Vartype.INTEGER, 'INTEGER', Vartype.REAL, 'REAL'], extended: bool = False) → Vartype

Cast various inputs to a valid vartype object.

Parameters:

• vartype (Vartype/str/set) –

  Variable type. Accepted input values:

  • SPIN, 'SPIN', {-1, 1}

  • BINARY, 'BINARY', {0, 1}

• extended (bool, optional, default=False) –

  If True, vartype can also be:

  • INTEGER, 'INTEGER'

  • REAL, 'REAL'

Returns:

Either SPIN or BINARY. If extended is True, can also be INTEGER or REAL

Return type:

Vartype

See also

vartype_argument()

Bound Information

vartype_info(vartype, dtype=<class 'numpy.float64'>)

Information about the variable bounds by variable type.

Parameters:

• vartype –

  Variable type. One of:

  • SPIN, 'SPIN', {-1, 1}

  • BINARY, 'BINARY', {0, 1}

  • INTEGER, 'INTEGER'

• dtype – One of float64 and float32.

Returns:

A named tuple with default_min, default_max, min, and max fields. These specify the default and largest bounds for the given variable type.