.. DO NOT EDIT.
.. THIS FILE WAS AUTOMATICALLY GENERATED BY SPHINX-GALLERY.
.. TO MAKE CHANGES, EDIT THE SOURCE PYTHON FILE:
.. "auto_examples/Basic-RAG/BasicRAG_FewShotPrompt.py"
.. LINE NUMBERS ARE GIVEN BELOW.

.. only:: html

    .. note::
        :class: sphx-glr-download-link-note

        :ref:`Go to the end <sphx_glr_download_auto_examples_Basic-RAG_BasicRAG_FewShotPrompt.py>`
        to download the full example code.

.. rst-class:: sphx-glr-example-title

.. _sphx_glr_auto_examples_Basic-RAG_BasicRAG_FewShotPrompt.py:

Custom Few-Shot Prompts
============================
This cookbook demonstrates how to use custom few-shot prompts with Basic RAG.

`Note that this cookbook assumes that you already have the` ``Llama-2-13b-chat`` `LLM ready, `
`for more details on how to quantize and run an LLM locally, `
`refer to the LLM section under Getting Started.`

`Note that this cookbook also assumes that you have already ingested documents into a DeepLake collection called 'grag'`
`for more details on how to ingest documents refer to the cookbook called` ``Document Ingestion``.

.. GENERATED FROM PYTHON SOURCE LINES 12-55

.. code-block:: Python


    from grag.components.multivec_retriever import Retriever
    from grag.components.prompt import FewShotPrompt
    from grag.components.vectordb.deeplake_client import DeepLakeClient
    from grag.rag.basic_rag import BasicRAG

    custom_few_shot_prompt = FewShotPrompt(
        input_keys={"context", "question"},
        output_keys={"answer"},
        example_template="""
        question: {question}
        answer: {answer}
        """,
        prefix="""Answer the following question based on the given context like examples given below:""",
        suffix="""Answer the following question based on the given context
        question: {question}
        context: {context}
        answer:
        """,
        examples=[
            {
                "question": "What is the name of largest planet?",
                "answer": "Jupiter is the largest planet.",
            },
            {
                "question": "Who came up with Convolutional Neural Networks?",
                "answer": "Yann LeCun introduced convolutional neural networks.",
            },
        ],
    )

    client = DeepLakeClient(collection_name="grag")
    retriever = Retriever(vectordb=client)
    rag = BasicRAG(
        model_name="Llama-2-13b-chat",
        custom_prompt=custom_few_shot_prompt,
        retriever=retriever,
    )

    if __name__ == "__main__":
        while True:
            query = input("Query:")
            rag(query)


.. _sphx_glr_download_auto_examples_Basic-RAG_BasicRAG_FewShotPrompt.py:

.. only:: html

  .. container:: sphx-glr-footer sphx-glr-footer-example

    .. container:: sphx-glr-download sphx-glr-download-jupyter

      :download:`Download Jupyter notebook: BasicRAG_FewShotPrompt.ipynb <BasicRAG_FewShotPrompt.ipynb>`

    .. container:: sphx-glr-download sphx-glr-download-python

      :download:`Download Python source code: BasicRAG_FewShotPrompt.py <BasicRAG_FewShotPrompt.py>`


.. only:: html

 .. rst-class:: sphx-glr-signature

    `Gallery generated by Sphinx-Gallery <https://sphinx-gallery.github.io>`_