| | |
- abc.ABC(builtins.object)
-
- StructureLearningProposal
class StructureLearningProposal(abc.ABC) |
| |
StructureLearningProposal(initial_state: ~State, blacklist=None, whitelist=None, seed: int = 32)
Base class for proposal classes for structure learning using MCMC.
All inheriting classes must implement the following methods:
propose() -> graph : numpy.ndarray, operation : str
compute_acceptance_ratio() -> float
|
| |
- Method resolution order:
- StructureLearningProposal
- abc.ABC
- builtins.object
Methods defined here:
- __init__(self, initial_state: ~State, blacklist=None, whitelist=None, seed: int = 32)
- Initialise StructureLearningProposal instance.
Parameters:
graph (networkx.DiGraph): graph
blacklist (numpy.ndarray): mask for edges to ignore in the proposal
whitelist (numpy.ndarray): mask for edges to include in the proposal
- accept(self)
- compute_acceptance_ratio(self, current_state_score, proposed_state_score, current_state_prior=0, proposed_state_prior=0) -> float
- get_nodes_to_rescore(self) -> List[str]
- propose(self) -> Tuple[~State, str]
- Propose a DAG
Data descriptors defined here:
- __dict__
- dictionary for instance variables
- __weakref__
- list of weak references to the object
Data and other attributes defined here:
- INITIAL = 'initial'
- STAY_STILL = 'stay_still'
- __abstractmethods__ = frozenset({'compute_acceptance_ratio', 'get_nodes_to_rescore', 'propose'})
- operations = ['stay_still']
| |