CommunityEmbedder

class convokit.hyperconvo.communityEmbedder.CommunityEmbedder(community_key: Optional[str] = None, n_components: int = 2, method: str = 'none')

Must be run after threadEmbedder.fit_transform()

Groups threads together into communities in this space for visualization or other such purposes.

Parameters
  • community_key – Key in “meta” dictionary of each utterance whose corresponding value we’ll use as the community label for that utterance (see threadEmbedder)

  • n_components – Number of dimensions to embed communities into

  • method – Embedding method; “svd”, “tsne” or “none”

fit_transform(corpus: convokit.model.corpus.Corpus) → convokit.model.corpus.Corpus

Groups threads together into communities.

Parameters

corpus – the Corpus to use

Returns

Modifies and returns Corpus with new meta key: “communityEmbedder”, value: Dict, containing “pts”: an array with rows corresponding to embedded communities, and “labels”: an array whose ith entry is the community of the ith row of X.

transform(corpus: convokit.model.corpus.Corpus) → convokit.model.corpus.Corpus

Same as fit_transform()