Module topic_model_extractor 

Topic Model Extractor — production-quality collapsed Gibbs sampling LDA.

Implements Latent Dirichlet Allocation (LDA) via collapsed Gibbs sampling for unsupervised topic discovery over text corpora. All randomness is driven by an xorshift64 PRNG so the implementation is 100 % pure-Rust with no rand dependency.

Structs

ExtractorConfig

Configuration for TopicModelExtractor.

ExtractorDocumentTopics

Per-document topic distribution produced by the extractor.

ExtractorTopic

A single latent topic produced by the extractor.

ExtractorTopicWord

A word and its probability / raw count within a topic.

ModelStats

Aggregate model statistics.

TopicModelExtractor

Production-quality collapsed Gibbs sampling LDA topic extractor.

Enums

ExtractorError

Errors that can be returned by TopicModelExtractor.

Type Aliases

TmeDocumentTopics

Type alias for ExtractorDocumentTopics — avoids collision with DocumentTopics from topic_modeler.

TmeError

Type alias for ExtractorError — convenience alias.

TmeTopic

Type alias for ExtractorTopic — convenience alias.

TmeTopicWord

Type alias for ExtractorTopicWord — avoids collision with TopicWord from topic_modeler.