lakeapi 0.22.2

API for accessing Lake crypto market data

Lake API

API for accessing Lake crypto market data.

Lake is a service providing historical cryptocurrency market data in high detail, including order book data, tick trades and 1m trade candles. It is tuned for convenient quant and machine-learning purposes and so offers high performance, caching and parallelization.

• Web: https://crypto-lake.com/

• Documentation: https://lake-api.readthedocs.io.

• Online example – executable collab notebook: https://colab.research.google.com/drive/1E7MSUT8xqYTMVLiq_rMBLNcZmI_KusK3

Usage

If you don’t have a paid plan with AWS credentials, you can access sample data:

import lakeapi

lakeapi.use_sample_data(anonymous_access = True)

df = lakeapi.load_data(
    table="book",
    start=None,
    end=None,
    symbols=["BTC-USDT"],
    exchanges=["BINANCE"],
)

With paid access, you can query any data:

import lakeapi

# Downloads SOL-USDT depth snapshots for last 2 days from Kucoin exchange
df = lakeapi.load_data(
    table="trades",
    start=datetime.datetime.now() - datetime.timedelta(days=2),
    end=None,
    symbols=["SOL-USDT"],
    exchanges=["KUCOIN"],
)

We recommend putting .lake_cache directory into .gitignore, because Lake API stores cache into this directory in the working directory.

History

0.22.1 (2025-05-15)

• fix list_data bug in last_modified_begin parameter

• check for crossed bests in orderbook.py on day boundaries

0.22.0 (2025-05-08)

• support multiple days in orderbook.py

0.21.0 (2025-05-04)

• fix joblib issue with unexpected keyword argument ‘bytes_limit’

0.20.1 (2025-04-28)

• fix UnboundLocalError in _get_table_contents_cache_key when non-existing table is passed

0.20.0 (2025-03-28)

• new used_data function

• clear cache on 404 to read new contents

0.19.1 (2024-12-16)

• fix typing issues

0.19.0 (2024-09-09)

• fix possible KeyError: ‘dt’ with list_data

0.18.0 (2024-08-21)

• fix KeyError: ‘dt’ with s3 method

0.17.0 (2024-08-09)

• orderbook class performance improvements

0.16.0 (2024-08-06)

• improve data loading performance, especially for the first time

0.15.0 (2024-07-24)

• fix conflict of new numpy 2 and older pyarrow

• improve tox tests

• add missing numba dependency needed by orderbook class (the rest works without numba)

0.14.0 (2023-05-02)

• order book helper class

0.13.0 (2023-03-15)

• fix old funding rates

0.12.1 (2023-11-22)

• fix issue !6 with lowercase symbol names

• better error when aws clients use free data sample (!7)

0.12.0 (2023-11-18)

• option to disable cache

• improved warnings handling

0.11.2 (2023-11-09)

• fix order book data loading (KeyError: side)

0.11.1 (2023-11-06)

• minor optimizations and fixes in cloudfront transfer

0.10.0 (2023-11-06)

• more efficient optional data transfer implemented via aws cloudfront

0.9.1 (2023-11-03)

• awswrangler dependency removed, pandas 2 support added

• python3.12 support

0.8.0 (2023-09-18)

• grow default cache size limit

• nicer error messages when data are missing

• pass and print warning when file is corrupted

0.7.0 (2023-09-18)

• let user specify max cache size via lakeapi.set_cache_size_limit()

0.6.4 (2023-08-05)

• too many open files bugfix

0.6.3 (2023-08-03)

• logging fixes

0.6.2 (2023-08-18)

• fix dependency constraints causing TypeError: _path2list() got an unexpected keyword argument ‘boto3_session’

0.6.1 (2023-08-15)

• fix path2list bug

• fix type hints

0.6.0 (2023-08-14)

• support for python3.11

0.5.0 (2023-05-21)

• support for funding, open_interest and liquidations list_data

• improve data type typing

0.4.5 (2023-01-09)

• grow default cache size limit from 3 GB to 10 GB

0.4.3 (2022-12-09)

• small documentation improvements

0.4.2 (2022-12-09)

• fix trades_mpid issue

0.4.1 (2022-12-05)

• fix warning messages in anonymous mode

0.4.0 (2022-11-19)

• level_1 data added to typing

• s3 user agent set to lakeapi

0.3.0 (2022-11-04)

• Typing bugfix

• Last modified filters for list_data

0.2.0 (2022-10-26)

• New feature for listing available data.

0.1.3 (2022-10-13)

• Corrupted cache bugfix

0.1.2 (2022-10-10)

• Caching and requirements improvements.

0.1.1 (2022-10-09)

• Python2.7 support and documentation improvements.

0.1.0 (2022-10-08)

• First release on PyPI.