strategy.py

import talib
import pandas as pd


class Strategy():
    def __init__(self, client, timeframe='5m'):
        self.client = client
        self.timeframe = timeframe

    def predict(self):
        ohlcv_candles = pd.DataFrame(self.client.Trade.Trade_getBucketed(
            binSize=self.timeframe,
            symbol='XBTUSD',
            count=100,
            reverse=True
        ).result()[0])

        ohlcv_candles.set_index(['timestamp'], inplace=True)

        macd, signal, hist = talib.MACD(ohlcv_candles.close.values,
                                        fastperiod=8, slowperiod=28, signalperiod=9)

        # sell
        if hist[-2] > 0 and hist[-1] < 0:
            return -1
        # buy
        elif hist[-2] < 0 and hist[-1] > 0:
            return 1
        # do nothing
        else:
            return 0