PYJUQUE     (pai-jook) (Py-thon Ju-ju Qu-ant E-ngine)

This project implements the basic functionality required to engage in algorithmic trading. It can be regarded as a starting point for more complex trading bots.

Make sure you have pip installed. Run:

# To install all the requirements
pip install -r requirements.txt

Add a .env file in the root filepath, then populate it with the following lines. Checkout this example .env file

BINANCE_API_KEY=...
BINANCE_API_SECRET=...

You should be good to go! Check out the example section.

Code is contained in pyjuque. Tests are in tests.

This library implements the following features:

At pyjuque/Strategies.

A base module which allows us to define buying & selling strategies for crypto assets. Each strategy will contain the following phases: setup (where the indicators are computed), getIndicators (to be used for plotting), checkLongSignal, checkShortSignal, getBuySignalsList and getSellSignalsList (the last two to be used for backtesting).

Currently contains a few basic strategies. More strategies will be added together with a build your own strategy template. Should seamlessly connect to the Backtesting & Bot Controller modules.

At pyjuque/Engine/BotController.py.

A module which handles the buying and selling of assets, given simple or more advanced rules, allowing us to run a strategy indefinitely. Checkout this example.

• Placing Entry (Buy) Orders on Signal
• Market, Limit & Stop Loss orders
• Placing Exit Order when Entry Order was fulfilled
• Trading below/above signal by some %

• OCO orders
• Selling on signals
• Trailing stop loss
• Multiple trade entries (in case trade goes against you)

State persistence Using SQLAlchemy.

At pyjuque/Engine/Backtester.

Backtester infrastructure currently supports:

• stop loss, trailing stop loss
• subsequent entries logic (as if Dollar Cost Averaging is enabled)

Checkout this example.

There is some basic functionality for plotting in pyjuque/Plotter

Implementing multiple exchanges with ccxt. Check out implementation at CcxtExchange. Currently implemented:

binance okex

Older:

At pyjuque/Exchanges.

• Binance - based on the official REST API

At pyjuque/Indicators.

Started implementing the Indicators module which currently contains some indicators from pyti. Undergoing integration with pandas_ta.

The thinking is that this module should allow us to easily and quickly compute any of the hundreds of indicators out there and to use them in strategies & backtesting. Should seamlessly connect to the Strategies module.

At pyjuque/Strategies/StrategyOptimiser.py.

Currently allows for optimising strategy parameters using a genetic algorithm. Checkout this example.

At pyjuque/Exchanges/BinanceOrderBook.py.

Creates and stores a local order book for the specified symbols. Order Book is updated every second through a websocket connection to the Exchange (currently Binance). Checkout this example.

from pyjuque.Exchanges.BinanceOrderBook import OrderBook

# Initialize & start OrderBook with desired symbols
ob = OrderBook(symbols=['BTCUSDT', 'LTCUSDT'])
ob.startOrderBook()
...
# Get Updated Order Book data at any point in your code 
ordb = ob.getOrderBook()
print(ordb)

{
  'BTCUSDT': {
      'asks': [
          ['13662.31000000', '3.24473100'],
          ['13662.82000000', '0.06815300'],
          ['13663.08000000', '0.00900000'],
          ...
          ['20000.00000000', '95.22325900']
        ],
      'bids': [
          ['13662.30000000', '1.26362900'],
          ['13661.78000000', '0.04395000'],
          ['13661.62000000', '0.01439200'],
          ...
          ['10188.00000000', '1.11546400']
        ],
      'lastUpdateId': 6382686192  # ignore this
  },
  'LTCUSDT': {
      'asks': [ ... ],
      'bids': [ ... ],
      'lastUpdateId': 1521585540  # ignore this
  },
 'counter': 11                    # ignore this
}

Run them with the command nose2

Binance Futures, Bitmex, Bitfinex, FTX, Bybit. Margin Trading, Market Making, Hyper Parameter Tuning.

To contribute simply fork the repo, write your desired feature in your own fork and make a pull request upon finishing. Please write tests!

Each exchange should extend the BaseExchange class and implement all functions there.