Quickstart

There are mainly 4 ways to interact with data in pfeed:

1. get_historical_data() gets historical data from local storage (if exists) or remote data sources and returns a cleaned DataFrame/LazyFrame.

import pfeed as pe
feed = pe.BybitFeed(data_tool='pandas')
df = feed.get_historical_data('BTC_USDT_PERP', rollback_period='1w', resolution='1d')

2. download_historical_data() or download() loads the downloaded data into your local machine, local data lake MinIO, or the cloud.

pe.download('bybit', products=['BTC_USDT_PERP'], dtypes=['minute'], start_date='2024-01-01', end_date='2024-01-03', use_ray=True, use_minio=False)

3. 🚧 get_realtime_data() gets real-time data by calling the broker/exchange’s API.

data: dict = feed.get_realtime_data(...)

4. 🚧 stream_realtime_data() or stream() listens to real-time data via websocket and optionally stores it .

pe.stream(...)