main.py

"""
    币安推荐码:  返佣10%
    https://www.binancezh.pro/cn/register?ref=AIR1GC70

    币安合约推荐码: 返佣10%
    https://www.binancezh.com/cn/futures/ref/51bitquant

    if you don't have a binance account, you can use the invitation link to register one:
    https://www.binancezh.com/cn/futures/ref/51bitquant

    or use the inviation code: 51bitquant

    风险提示: 网格交易在单边行情的时候，会承受比较大的风险，请你了解整个代码的逻辑，然后再使用。
    RISK NOTE: Grid trading will endure great risk at trend market, please check the code before use it. USE AT YOUR OWN RISK.

"""

import time
import logging
from trader.binance_spot_trader import BinanceSpotTrader
from trader.binance_future_trader import BinanceFutureTrader
from utils import config
from apscheduler.schedulers.background import BackgroundScheduler

format = '%(asctime)s - %(name)s - %(levelname)s - %(message)s'
logging.basicConfig(level=logging.INFO, format=format, filename='log.txt')
logging.getLogger("apscheduler.scheduler").setLevel(logging.WARNING)
logging.getLogger("apscheduler.executors.default").setLevel(logging.WARNING)

logger = logging.getLogger('binance')
from typing import Union
from gateway.binance_future import Interval
import numpy as np
import pandas as pd
from datetime import datetime

pd.set_option('expand_frame_repr', False)

from utils.config import signal_data


def get_data(trader: Union[BinanceFutureTrader, BinanceSpotTrader]):
    # traders.symbols is a dict data structure.
    symbols = trader.symbols_dict.keys()

    signals = []

    # we calculate the signal here.
    if len(config.allowed_lists) > 0:
        symbols = config.allowed_lists

    for symbol in symbols:

        if len(config.blocked_lists) > 0:
            if symbol.upper() in config.blocked_lists:
                continue

        klines = trader.get_klines(symbol=symbol.upper(), interval=Interval.HOUR_1, limit=100)
        if len(klines) > 0:
            df = pd.DataFrame(klines, dtype=np.float64,
                              columns=['open_time', 'open', 'high', 'low', 'close', 'volume', 'close_time', 'turnover', 'a2',
                                       'a3', 'a4', 'a5'])
            df = df[['open_time', 'open', 'high', 'low', 'close', 'volume', 'turnover']]
            df.set_index('open_time', inplace=True)
            df.index = pd.to_datetime(df.index, unit='ms') + pd.Timedelta(hours=8)

            df_4hour = df.resample(rule='4H').agg({'open': 'first',
                                        'high': 'max',
                                        'low': 'min',
                                        'close': 'last',
                                        'volume': 'sum',
                                        'turnover': 'sum'
                                        })

            # print(df)

            # calculate the pair's price change is one hour. you can modify the code below.
            pct = df['close'] / df['open'] - 1
            pct_4h = df_4hour['close']/df_4hour['open'] - 1

            value = {'pct': pct.iloc[-1], 'pct_4h':pct_4h.iloc[-1] , 'symbol': symbol, 'hour_turnover': df['turnover'].iloc[-1]}


            # calculate your signal here.
            if value['pct'] >= config.pump_pct or value['pct_4h'] >= config.pump_pct_4h:
                # the signal 1 mean buy signal.
                value['signal'] = 1
            elif value['pct'] <= -config.pump_pct or value['pct_4h'] <= -config.pump_pct_4h:
                value['signal'] = -1
            else:
                value['signal'] = 0

            signals.append(value)

    signals.sort(key=lambda x: x['pct'], reverse=True)
    signal_data['id'] = signal_data['id'] + 1
    signal_data['time'] = datetime.now()
    signal_data['signals'] = signals
    print(signal_data)

if __name__ == '__main__':

    config.loads('./config.json')
    print(config.blocked_lists)

    if config.platform == 'binance_spot':
        # if you want to trade spot, set the platform to 'binance_spot',  else will trade Binance Future(USDT Base)
        # 如果你交易的是币安现货，就设置config.platform 为 'binance_spot'，否则就交易的是币安永续合约(USDT)
        trader = BinanceSpotTrader()
    else:
        trader = BinanceFutureTrader()


    trader.get_exchange_info()
    get_data(trader)  # for testing

    scheduler = BackgroundScheduler()
    scheduler.add_job(get_data, trigger='cron', hour='*/1', args=(trader,))
    scheduler.start()

    while True:
        time.sleep(10)
        trader.start()

"""
策略逻辑: 

1. 每1个小时会挑选出前几个波动率最大的交易对(假设交易的是四个交易对).
2. 然后根据设置的参数进行下单(假设有两个仓位,那么波动率最大的两个，且他们过去一段时间是暴涨过的)
3. 然后让他们执行马丁策略.


Martingle trading strategy: 

1. select the top trading pairs with the highest volatility every hour (assuming four trading pairs are traded, you can config in the config.json file)

2. Then place an order based on the setting parameters

3. Then have them execute the Martin strategy

"""