Data Science (Pandas)

If you are performing data analysis, plotting telemetry with matplotlib, or training machine learning models, raw Python lists of Pydantic objects can be slow and cumbersome.

pyopenf1 includes a built-in extension to instantly flatten these payloads into high-performance pandas DataFrames.

Installation

Ensure you have the pandas extra installed:

pip install "pyopenf1[pandas]"

Creating DataFrames

Use the to_dataframe() utility function to convert any list of OpenF1 models into a DataFrame.

import asyncio
from pyopenf1 import AsyncOpenF1Client
from pyopenf1.ext.pandas import to_dataframe

async def analyze_telemetry():
    async with AsyncOpenF1Client() as client:
        # 1. Fetch raw Pydantic models
        raw_data = await client.telemetry.get_car_data(driver_number=1, session_key=9159)

        # 2. Convert to DataFrame
        df = to_dataframe(raw_data)

        # 3. Analyze!
        print(df.describe())
        print(f"Max Speed: {df['speed'].max()} km/h")

asyncio.run(analyze_telemetry())

[!TIP] to_dataframe automatically handles the timezone-aware datetime objects parsed by Pydantic, ensuring your DataFrame indices and time-series plots work flawlessly out of the box!