Field Labeled CSV Reader

The Field Labeled CSV Reader converts CSV rows into field-labeled text documents, making them more readable for natural language processing and agent-based retrieval systems.

Code

cookbook/knowledge/readers/csv_field_labeled_reader.py

from agno.agent import Agent
from agno.knowledge.knowledge import Knowledge
from agno.knowledge.reader.field_labeled_csv_reader import FieldLabeledCSVReader
from agno.vectordb.pgvector import PgVector

db_url = "postgresql+psycopg://ai:ai@localhost:5532/ai"

reader = FieldLabeledCSVReader(
    chunk_title="🎬 Movie Information",
    field_names=[
        "Movie Rank",
        "Movie Title",
        "Genre",
        "Description",
        "Director",
        "Actors",
        "Year",
        "Runtime (Minutes)",
        "Rating",
        "Votes",
        "Revenue (Millions)",
        "Metascore",
    ],
    format_headers=True,
    skip_empty_fields=True,
)

knowledge_base = Knowledge(
    vector_db=PgVector(
        table_name="imdb_movies_field_labeled_readr",
        db_url=db_url,
    ),
)

knowledge_base.add_content(
    url="https://agno-public.s3.amazonaws.com/demo_data/IMDB-Movie-Data.csv",
    reader=reader,
)

agent = Agent(
    knowledge=knowledge_base,
    search_knowledge=True,
    instructions=[
        "You are a movie expert assistant.",
        "Use the search_knowledge_base tool to find detailed information about movies.",
        "The movie data is formatted in a field-labeled, human-readable way with clear field labels.",
        "Each movie entry starts with '🎬 Movie Information' followed by labeled fields.",
        "Provide comprehensive answers based on the movie information available.",
    ],
)

agent.print_response(
    "which movies are directed by Christopher Nolan", markdown=True, stream=True
)

Usage

1. Create a virtual environment

   Open the Terminal and create a python virtual environment.

   Mac

   python3 -m venv .venv
   source .venv/bin/activate
   
2. Install libraries

   pip install -U agno aiofiles sqlalchemy psycopg[binary] pgvector openai
   
3. Set environment variables

   export OPENAI_API_KEY=xxx
   
4. Run PgVector

   docker run -d \
     -e POSTGRES_DB=ai \
     -e POSTGRES_USER=ai \
     -e POSTGRES_PASSWORD=ai \
     -e PGDATA=/var/lib/postgresql/data/pgdata \
     -v pgvolume:/var/lib/postgresql/data \
     -p 5532:5432 \
     --name pgvector \
     agno/pgvector:16
   
5. Run Agent

   Mac

   python cookbook/knowledge/readers/csv_field_labeled_reader.py
   

Params

Parameter	Type	Default	Description
file	Union[Path, IO[Any]]	Required	Path to CSV file or file-like object
chunk_title	Optional[Union[str, List[str]]]	None	Title to add at the top of each entry. Can be a single string or a list that rotates through entries
field_names	Optional[List[str]]	[]	Custom labels for CSV fields. If not provided, column headers are used
format_headers	bool	True	Whether to format column headers by replacing underscores with spaces and applying title case
skip_empty_fields	bool	True	Whether to skip fields with empty values in the output
delimiter	str	","	Character used to separate fields in the CSV
quotechar	str	'"'	Character used to quote fields in the CSV
encoding	Optional[str]	"utf-8"	Character encoding for reading the file