Welcome to binance-history’s documentation!#

binance-history is a Python library for fetching cryptocurrency klines or trades data. It downloads data from the Binance Public Data.

API#

binance_history.fetch_klines(symbol, start, end, timeframe='1m', asset_type='spot', tz=None)#
Parameters:

  • symbol (str) – The binance market pair name. e.g. 'BTCUSDT'.

  • start (str | datetime) – The start datetime of requested data. If it’s an instance of datetime.datetime, it’s timezone is ignored. If it’s a str, it should be parsed by dateutil, e.g. "2022-1-1 8:10".

  • end (str | datetime) –

    The end datetime of requested data. If it’s an instance of datetime.datetime, it’s timezone is ignored. If it’s a str, it should be parsed by dateutil, e.g. "2022-1-2 8:10".

  • timeframe (str) – The kline interval. e.g. “1m”. see binance_history.constants.TIMEFRAMES to see the full list of available intervals.

  • asset_type (str) – The asset type of requested data. It must be one of 'spot', 'futures/um', 'futures/cm'.

  • tz (str | None) – Timezone of start, end, and the open/close datetime of the returned dataframe. It should be a time zone name of tz database, e.g. “Asia/Shanghai”. Your can find a full list of available time zone names in List of tz database time zones.

Returns:

A pandas dataframe with columns open, high, low, close, volume, trades, close_datetime. the dataframe’s index is the open datetime of klines, the timezone of the datetime is set by tz_database_name, if it is None, your local timezone will be used.

Return type:

DataFrame

binance_history.fetch_agg_trades(symbol, start, end, asset_type='spot', tz=None)#
Parameters:

  • symbol (str) – The binance market pair name. e.g. 'BTCUSDT'.

  • start (str | datetime) –

    The start datetime of requested data. If it’s an instance of datetime.datetime, it’s timezone is ignored. If it’s a str, it should be parsed by dateutil, e.g. "2022-1-1 8:10".

  • end (str | datetime) –

    The end datetime of requested data. If it’s an instance of datetime.datetime, it’s timezone is ignored. If it’s a str, it should be parsed by dateutil, e.g. "2022-1-2 8:10".

  • asset_type (str) – The asset type of requested data. It must be one of 'spot', 'futures/um', 'futures/cm'.

  • tz (str | None) –

    Timezone of start, end, and the open/close datetime of the returned dataframe. It should be a time zone name of tz database, e.g. “Asia/Shanghai”. Your can find a full list of available time zone names in List of tz database time zones.

Returns:

A pandas dataframe with columns price, quantity, is_buyer_maker, the dataframe’s index is the datetime of the aggregated trades, the timezone of the datetime is set by tz_database_name, if it is None, your local timezone will be used.

Return type:

DataFrame