Teradata Package for LangChain Function Reference - IngestParams - Teradata® Package for LangChain - Look here for syntax, methods and examples for the functions included in the Teradata langchain-teradata package.

libs.teradata.langchain_teradata.IngestParams = class IngestParams(builtins.object)

libs.teradata.langchain_teradata.IngestParams(**kwargs)

Class sets up the ingestor-related parameters for vector store creation.

Methods defined here:

__init__(self, **kwargs): DESCRIPTION:
    Instantiates the IngestParams class to set up the
    ingestor-related parameters for vector store creation.
    Note:
        * Only applicable for file-based vector store.
        * Only applicable for AI-Factory system.

PARAMETERS:
    nv_ingestor:
        Optional Argument.
        Specifies whether to use NVIDIA NV-Ingest for processing the
        document files.
        Notes:
            * Applicable only while using NVIDIA NIM endpoints.
            * Applicable only for 'file-based' vector stores.
        Types: bool

    ingest_host:
        Required for NVIDIA NIM, Optional otherwise.
        Specifies the HTTP host for the service to be used for document parsing.
        Note:
            * Applicable only while using NVIDIA NIM endpoints.
        Types: str

    ingest_port:
        Optional Argument.
        Specifies the HTTP port for the service to be used for document parsing.
        Note:
            * Applicable only while using NVIDIA NIM endpoints.
        Default Value: 7670
        Types: int

    display_metadata:
        Optional Argument.
        Specifies whether to display metadata describing objects extracted
        from document files when using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Types: bool

    extract_text:
        Optional Argument.
        Specifies whether to extract text from the document files when
        using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Default Value: True
        Types: bool

    extract_images:
        Optional Argument.
        Specifies whether to extract images from the document files when
        using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Default Value: True
        Types: bool

    extract_tables:
        Optional Argument.
        Specifies whether to extract tables from the document files when
        using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Default Value: True
        Types: bool

    extract_infographics:
        Required for NVIDIA NIM, Optional otherwise.
        Specifies whether to extract infographics from
        document files.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Types: bool

    extract_method:
        Required for NVIDIA NIM, Optional otherwise.
        Specifies the method to be used for extracting text from
        the document files.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Default Value: pdfium
        Permitted Values: pdfium, nemoretriever_parse
        Types: str

    extract_metadata_json:
        Optional Argument.
        Specifies whether to extract metadata in JSON format and store it in
        the metadata_json column when using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Types: bool

    extract_caption:
        Optional Argument.
        Specifies whether to extract captions for images and tables from
        the document files when using NVIDIA NV-Ingest.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints and
              when "nv_ingestor" is set to True.
        Types: bool

    tokenizer:
        Optional Argument.
        Specifies the tokenizer to be used for splitting the text into chunks.
        Notes:
            * Applicable only when "nv_ingestor" is set to True
              and "document_files" is supplied.
            * Applicable only while using NVIDIA NIM endpoints.
        Default Value: meta-llama/Llama-3.2-1B
        Types: str

    overwrite_object:
        Optional Argument.
        Specifies whether to overwrite the existing object with
        the same name in the database.
        Note:
            * Applicable only for 'file-based' vector stores.
        Types: bool

    vlm_base_url:
        Optional Argument.
        Specifies the base URL for Vision Language Model when extract_caption from images is enabled.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints.
            * Applicable only when "extract_caption" is set to True.
        Types: str

    vlm_model:
        Optional Argument.
        Specifies the Vision Language Model to be used when extract_caption from images is enabled.
        Notes:
            * Applicable only for 'file-based' vector stores.
            * Applicable only while using NVIDIA NIM endpoints.
            * Applicable only when "extract_caption" is set to True.
        Types: str

RETURNS:
    None.

RAISES:
    TeradataMlException.

EXAMPLES:
# Import the required class
>>> from teradatagenai import IngestParams

# Set up ingestor parameters
>>> ingest_params = IngestParams(nv_ingestor = True,
                                 ingest_host = "<Your ingestor host>",
                                 ingest_port = 443,
                                 extract_text = True,
                                 extract_images = True)

Data descriptors defined here:

__dict__: dictionary for instance variables

__weakref__: list of weak references to the object