PDF KnowledgeBase - Phidata

Example

Install pypdf if needed using this guide

knowledge_base.py

from phi.knowledge.pdf import PDFKnowledgeBase, PDFReader
from phi.vectordb.pgvector import PgVector

from resources import vector_db

pdf_knowledge_base = PDFKnowledgeBase(
    path="data/pdfs",
    # Table name: llm.pdf_documents
    vector_db=PgVector(
        table_name="pdf_documents",
        db_url=vector_db.get_db_connection_local(),
    ),
    reader=PDFReader(chunk=True),
)

PDFKnowledgeBase Params

Parameter	Type	Default	Description
`path`	`Union[str, Path]`	-	Path to `PDF` files. Can point to a single PDF file or a directory of PDF files.
`reader`	`Union[PDFReader, PDFImageReader]`	`PDFReader()`	A `PDFReader` or `PDFImageReader` that converts the `PDFs` into `Documents` for the vector database.

AgentKnowledge Params

PDFKnowledgeBase is a subclass of the AgentKnowledge class and has access to the same params

Parameter	Type	Default	Description
`reader`	`Optional[Reader]`	`None`	Reader to read the documents
`vector_db`	`Optional[VectorDb]`	`None`	Vector db to store the knowledge base
`num_documents`	`int`	`2`	Number of relevant documents to return on search
`optimize_on`	`Optional[int]`	`1000`	Number of documents to optimize the vector db on
`driver`	`str`	`"knowledge"`	Driver for the Assistant knowledge