SemanticSearchSchema

Schema for semantic search

Subclass of Executable. Type of Executable[SemanticSearchSchemaInput, SemanticSearchSchemaOutput]

Module: implementation.schemas

Methods and properties

Main methods and properties

---

__init__

Arguments:

• dataset (Optional[List[str]], optional): Dataset for search. Defaults to None.

• encoder (Optional[TransformersTextEmbedding[Any, Any]], optional): Encoder for embeddings creation. If equals to None, default encoder will be used. Defaults to None.

• input_class (Type[SemanticSearchSchemaInput], optional): Class for input validation. Defaults to SemanticSearchSchemaInput.

• output_class (Type[SemanticSearchSchemaOutput], optional): Class for output validation. Defaults to SemanticSearchSchemaOutput.

• name (Optional[str], optional): Name for identification. If equals to None, class name will be used. Defaults to None.

---

---

---

SemanticSearchSchemaInput

Subclass of IOModel.

---

__init__

Arguments:

• query (List[str])

• results_count (int)

---

---

---

SemanticSearchSchemaOutput

Subclass of IOModel.

---

__init__

Arguments:

• search_results (Dict[str, Any]): Expected keys:

  • "distances" (List[float]): List of distances;

  • "indexes" (List[Any]): Indexes of indexed data;

  • "texts" (List[str]): Texts of results;

---

---