josiann.moves.discrete.SetStretch
- class josiann.moves.discrete.SetStretch(*, position_set, a=2.0, bounds=None, repr_attributes=(), **kwargs)[source]
Fusion of the Set and Stretch moves. We exploit multiple walkers in parallel a move each to the closest point in the set of possible positions instead of the point proposed by the stretch.
Instantiate a Move.
- Parameters:
position_set (
Sequence
[Sequence
[float
]]) – sets of only possible values for x in each dimension.a (
float
(default:2.0
)) – parameter for tuning the distribution of Z. Smaller values make samples tightly distributed around 1 while bigger values make samples more spread out with a peak getting closer to 0.bounds (
Optional
[ndarray
[Any
,dtype
[Union
[float64
,int64
]]]] (default:None
)) – optional sequence of (min, max) bounds for values to propose in each dimension.repr_attributes (
tuple
[str
,...
] (default:()
)) – tuple of attribute names to include in the move’s representation.kwargs (Any) –
Methods
Instantiate a Move.
Generate a new proposed vector x.
Set bounds for the move.
Methods
- SetStretch.__init__(*, position_set, a=2.0, bounds=None, repr_attributes=(), **kwargs)[source]
Instantiate a Move.
- Parameters:
position_set (
Sequence
[Sequence
[float
]]) – sets of only possible values for x in each dimension.a (
float
(default:2.0
)) – parameter for tuning the distribution of Z. Smaller values make samples tightly distributed around 1 while bigger values make samples more spread out with a peak getting closer to 0.bounds (
Optional
[ndarray
[Any
,dtype
[Union
[float64
,int64
]]]] (default:None
)) – optional sequence of (min, max) bounds for values to propose in each dimension.repr_attributes (
tuple
[str
,...
] (default:()
)) – tuple of attribute names to include in the move’s representation.kwargs (Any) –
- SetStretch.get_proposal(x, state)
Generate a new proposed vector x.