GLiNERQandA

Q&A task

Default predictor

This task uses GliNERPredictor by default.

GLiNERPredictor(
    GLiNERPredictorConfig(model_name="knowledgator/gliner-multitask-large-v0.5")
)

For more details, see:

GLiNERPredictor

By default, the predictor uses the knowledgator/gliner-multitask-large-v0.5 model. Learn more about the model here:

knowledgator/gliner-multitask-large-v0.5 · Hugging Facehuggingface

Subclass of NERTask.

Module: implementation.tasks

Methods and properties

Main methods and properties

---

__init__

Arguments:

• predictor (Predictor[Any, Any], optional): Predictor that will be used in task. If equals to None, default predictor will be used. Defaults to None.
• preprocess (Optional[Component], optional): Component executed before predictor.

  If equals to None, default component will be used. Defaults to None. Default component: GLiNERQandAPreprocessor
• postprocess (Optional[Component], optional): Component executed after predictor.

  If equals to None, default component will be used. Defaults to None. Default component: GLiNERQandAPostprocessor
• input_class (Type[Input], optional): Class for input validation. Defaults to GLiNERQandAInput.
• output_class (Type[Output], optional): Class for output validation. Defaults to GLiNERQandAOutput.
• name (Optional[str], optional): Name for identification. If equals to None, class name will be used. Defaults to None.

---

---

---

GLiNERQandAInput

Subclass of IOModel.

---

__init__

Arguments:

• text (str): Text to use.

• question (str): Question to answer.

---

---

---

GLiNERQandAOutput

Subclass of NEROutput. Type of NEROutput[Entity].

---

__init__

Arguments:

• text (str): Input text.

• question (str): Answered question.

• output (List[Entity]): Answers.

---

---

---

GLiNERQandAPreprocessor

Preprocess inputs. Subclass of Action. Type of Action[Dict[str, Any], Dict[str, Any]].

---

__init__

Arguments:

• sents_batch (int): Chunks size in sentences. Defaults to 10.

• threshold (float): Minimial score to put entities into the output (used by predictor). Defaults to 0.5.

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

---

execute

Arguments:

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

  • "text" (str): Text to process;

  • "question" (str): Question to answer;

Returns:

• Dict[str, Any]: Expected keys:

  • "texts" (List[str]): Model inputs;

  • "labels" (List[str]): Labels model inputs;

  • "chunks_starts" (List[int]): Chunks start positions. Used by postprocessor;

  • "threshold" (float): Minimal score for an entity to put into output;

---

---

---

GLiNERQandAPostprocessor

Format output. Subclass of Action. Type of Action[Dict[str, Any], Dict[str, Any]].

---

execute

Arguments:

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

  • "output" (List[List[Dict[str, Any]]]): Model output;

  • "chunks_starts" (List[int]): Chunks starts;

  • "text" (str): Processed text;

  • "question" (str): Answered question.

Returns:

• Dict[str, Any]: Expected keys:

  • "text" (str): Processed text;

  • "question" (str): Answered question.

  • "output" (List[Entity]): Answers;

---

---