Examples#
I’ve included three example highlighting the three main uses for the library: (1) Data analysis, (2) Building and maintaining datasets, and (3) Leverage statistical research into algorithmic trading systems. The GitHub distribution of this package includes an examples folder containing the .py files for each of the three below examples.
A Quick Data Example#
At the core of Kucoin-Cli, or KCI, is data acquisition functionality. While there are numerous other REST wrappers and order execution/management libraries available, none match the level of out-of-box detail paid by KCI. Below we will go through a quick code snippet showcasing some of the more important research functions the library has to offer as well as conducting a quick analysis on live data. I strongly encourage readers to skim the below code block and notes, then visit the examples folder where there is a Jupyter Notebook containing the same code in deeper detail. Note that this example was built on KCI version 1.4.6 or greater. Some functions may not run on older versions of the library.
Data Pipeline Setup#
Below is a simple code example describing how to setup a SQL database using kucoincli.pipe module. In kucoin-cli distributions later than 1.0.0, this code is included in the examples folder labels pipe_example.py.
""""
##### A simple data pipeline using `kucoincli.pipe` module #####
Database created will look like this ......
+----------+-------+-------+-------+-------+-------------+
| time | open | close | high | low | volume |
+----------+-------+-------+-------+-------+-------------+
|2022-01-01|1001.51|1002.21|1008.32| 999.43|8.14505485512|
|2022-01-02| 999.49|1000.80|1004.89| 995.32|9.15848158419|
...
...
...
"""
# Import our modules
from kucoincli.pipe import pipeline
from sqlalchemy import create_engine
import logging
# Create our sqlite database engine with sqlalchemy.
# The engine will generate a new database or append
# to a pre-existing one.
engine = create_engine("sqlite:///example.db")
# Add a logger to see pick up some additional output info
# To retrieve timeout messages set logging level to DEBUG
fmt = "%(asctime)s [%(levelname)s] %(module)s :: %(message)s"
logging.basicConfig(level=logging.INFO, format=fmt)
logging.getLogger(__name__)
# Setup constants for the pipeline.
# `pipeline`s are highly configurable
# Read the docs for more information
TICKER = "BTC-USDT"
START = "2022-04-01"
END = "2022-05-01"
INTERVAL = "1min"
# Now let's open up our pipeline ...
pipeline(
tickers=TICKER, # Tickers to query OHLCV data for
engine=engine, # Engine to run our database
interval=INTERVAL, # Interval at which to obtain OHLCV
start=START, # Earliest date to query from
end=END, # Latest date to query from
)
# Viola, we have generated a database.
# This current pipeline will obtain 30 days of minutely data for our `TICKER` const
# and store it in an SQLite .db file in our examples folder.
# Default it will track its progress with a progress bar and let the user know
# when it has finished retrieving and filing the data.
## Thats the entire pipeline! As easy as that we have created a permanent ##
## SQLite database to draw from for future research. See the SQLalchemy ##
## documentation (https://docs.sqlalchemy.org/en/14/) for details on how ##
## to interact with our new database. ##