Method analyse_query

document_qa/document_qa_engine.py:393–436

Compute a relevance coefficient for *query* against *doc_id*. The coefficient is ``min_similarity - mean_similarity`` over the top-k retrieved chunks. A value close to zero suggests the question matches multiple passages equally well. Args: query: The n

(self, query, doc_id, context_size=4)

        return relevant_documents

    def analyse_query(self, query, doc_id, context_size=4):
        """Compute a relevance coefficient for *query* against *doc_id*.

        The coefficient is ``min_similarity - mean_similarity`` over the
        top-k retrieved chunks.  A value close to zero suggests the
        question matches multiple passages equally well.

        Args:
            query: The natural-language question.
            doc_id: Document identifier.
            context_size: Number of chunks to consider (default 4).

        Returns:
            tuple: ``(summary_string, coordinates)``
        """
        db = self.data_storage.embeddings_dict[doc_id]
        # retriever = db.as_retriever(
        #     search_kwargs={"k": context_size, 'score_threshold': 0.0},
        #     search_type="similarity_score_threshold"
        # )
        retriever = db.as_retriever(search_kwargs={"k": context_size}, search_type="similarity_with_embeddings")
        relevant_documents = retriever.invoke(query)
        relevant_document_coordinates = [
            doc.metadata["coordinates"].split(";") if "coordinates" in doc.metadata else [] for doc in relevant_documents
        ]
        all_documents = db.get(include=["documents", "metadatas", "embeddings"])
        # all_documents_embeddings = all_documents["embeddings"]
        # query_embedding = db._embedding_function.embed_query(query)

        # distance_evaluator = load_evaluator("pairwise_embedding_distance",
        #                               embeddings=db._embedding_function,
        #                               distance_metric=EmbeddingDistance.EUCLIDEAN)

        # distance_evaluator.evaluate_string_pairs(query=query_embedding, documents="")

        similarities = [doc.metadata["__similarity"] for doc in relevant_documents]
        min_similarity = min(similarities)
        mean_similarity = sum(similarities) / len(similarities)
        coefficient = min_similarity - mean_similarity

        return (
            f"Coefficient: {coefficient}, (Min similarity {min_similarity}, Mean similarity: {mean_similarity})",
            relevant_document_coordinates,
        )

    def _parse_json(self, response, output_parser):
        system_message = (