nlg_generative_lm
=================

.. py:module:: nlg_generative_lm

.. autoapi-nested-parse::

   Generative natural language generation using a language model.



Classes
-------

.. autoapisummary::

   nlg_generative_lm.LMGenerativeNLG


Module Contents
---------------

.. py:class:: LMGenerativeNLG(ollama_config_file: str, prompt_file: str, prompt_prefix: Optional[str] = None)

   Bases: :py:obj:`dialoguekit.nlg.nlg_abstract.AbstractNLG`


   Initializes the generative NLG.

   :param ollama_config_file: Path to the Ollama configuration file.
   :param prompt_file: Path to the prompt file.
   :param prompt_prefix: Prefix to be remove from generated utterances.
                         Defaults to None.

   :raises FileNotFoundError: If the Ollama configuration file or the prompt
       file is not found.


   .. py:method:: generate_utterance_text(dialogue_acts: List[dialoguekit.core.dialogue_act.DialogueAct], annotations: Optional[Union[List[dialoguekit.core.annotation.Annotation], None]] = None, force_annotation: bool = False) -> Union[dialoguekit.core.annotated_utterance.AnnotatedUtterance, bool]

      Turns a structured utterance into a textual one.

      :param dialogue_acts: Dialogue acts of the utterance to be generated.
      :param annotations: If provided, these annotations should be considered
                          during generation.
      :param force_annotation: A flag to indicate whether annotations should be
                               forced or not. Not used in this NLG.

      :raises RuntimeError: If generation fails.

      :returns: Natural language utterance.