civiqhub_selene.blocks

Co-clustering of voters and items. Needs NumPy, behind the blocks extra.

A group gathers voters, a block gathers items, and a co-cluster is the pair of a group and a block. This subpackage holds the data the block models work on, the models, the choice of their shape and the consensus between groups.

Install it with pip install "civiqhub-selene[blocks]". Without NumPy, importing this subpackage raises MissingExtraError; the package root and civiqhub_selene.ordinal never import it.

Exported names

These names are imported from civiqhub_selene.blocks. Each one links to the module that defines it. The other names of co-clustering are imported from their module, listed below.

Name

What it is

Cocluster

One co-cluster: how a group of voters answers a block of items.

CoclusterTable

The table of the co-clusters of Q groups and L blocks.

GroupedVotes

A vote matrix read through groups of voters and blocks of items.

ItemBlock

A block: a cluster of items.

LatentBlockModel

Co-clustering where each voter belongs to one group and each item to one block.

OrdinalScale

An ordered scale of scores. Defaults to the five-score agreement scale.

Partition

Identifiers arranged in clusters. A sequence of its clusters.

Vote

One vote: a voter gave a score to an item.

VoteMatrix

Votes of M voters on P items, by opaque identifiers.

VoterGroup

A group: a cluster of voters.

Modules