relations
¶
Functions for computing relations between concepts.
- pyphi.relations.all_are_equal(seq='__no__default__')¶
- pyphi.relations.all_are_identical(seq='__no__default__')¶
- pyphi.relations.all_minima(seq='__no__default__')¶
Return the extrema of
seq
.Use
<
as the comparison to obtain the minima; use>
as the comparison to obtain the maxima.Uses only one pass through
seq
.- Parameters
comparison (callable) – A comparison operator.
seq (iterator) – An iterator over a sequence.
- Returns
The maxima/minima in
seq
.- Return type
list
- pyphi.relations.all_maxima(seq='__no__default__')¶
Return the extrema of
seq
.Use
<
as the comparison to obtain the minima; use>
as the comparison to obtain the maxima.Uses only one pass through
seq
.- Parameters
comparison (callable) – A comparison operator.
seq (iterator) – An iterator over a sequence.
- Returns
The maxima/minima in
seq
.- Return type
list
- pyphi.relations.indices(iterable)¶
Convert an iterable to element indices.
- pyphi.relations.divergence(p, q)¶
- pyphi.relations.maximal_state(mice)¶
Return the maximally divergent state(s) for this MICE.
Note that there can be ties.
- Returns
A 2D array where each row is a maximally divergent state.
- Return type
np.array
- pyphi.relations.congruent_nodes(states)¶
Return the set of nodes that have the same state in all given states.
- class pyphi.relations.Relation(relata, purview, phi, ties=None)¶
A relation among causes/effects.
- property relata¶
- property purview¶
- property phi¶
- property ties¶
- property subsystem¶
- property mechanisms¶
- order_by()¶
Return a list of values to compare for ordering.
The first value in the list has the greatest priority; if the first objects are equal the second object is compared, etc.
- static union(tied_relations)¶
Return the ‘union’ of tied relations.
This is a new Relation object that contains the purviews of the other relations in the
ties
attribute.
- class pyphi.relations.Relata(subsystem, relata)¶
A set of potentially-related causes/effects.
- property subsystem¶
- property mechanisms¶
- property purviews¶
- property maximal_states¶
- __iter__()¶
Iterate over relata.
- overlap()¶
Return the set of elements that are in the purview of every relatum.
- null_relation(purview=None, phi=0.0)¶
- congruent_overlap()¶
Yield the congruent overlap(s) among the relata.
These are the common purview elements among the relata whose maximally-divergent states are consistent; that is, the largest subset of the union of the purviews such that, for each element, that element’s state is the same according to the maximally divergent state of each relatum.
Note that there can be multiple congruent overlaps.
- possible_purviews()¶
Return all possible purviews.
This is the powerset of the congruent overlap. If there are multiple congruent overlaps because of ties, it is the union of the powerset of each.
- partitioned_divergence(purview, mice)¶
Return the maximal partitioned divergence over this purview.
The purview is cut away from the MICE and the divergence is computed between the unpartitioned repertoire and partitioned repertoire.
If the MICE has multiple tied maximally-divergent states, we take the maximum unpartitioned-partitioned divergence across those tied states.
- Parameters
purview (set) – The set of node indices in the purview.
mice (
MaximallyIrreducibleCauseOrEffect
) – TheMaximallyIrreducibleCauseOrEffect
object to consider.
- minimum_information_relation(purview)¶
Return the minimal information relation for this purview.
- Parameters
relata (Relata) – The relata to consider.
purview (set) – The purview to consider.
- maximally_irreducible_relation()¶
Return the maximally-irreducible relation among these relata.
If there are ties, the tied relations will be recorded in the ‘ties’ attribute of the returned relation.
- Returns
the maximally irreducible relation among these relata, with any tied purviews recorded.
- Return type
- pyphi.relations.relation(relata)¶
Return the maximally irreducible relation among the given relata.
Alias for the
Relata.maximally_irreducible_relation()
method.
- pyphi.relations.separate_ces(ces)¶
Return the individual causes and effects, unpaired, from a CES.
- pyphi.relations.all_relations(subsystem, ces)¶
Return all relations, even those with zero phi.
- pyphi.relations.relations(subsystem, ces)¶
Return the irreducible relations among the causes/effects in the CES.