从性能方面对策略进行优化的方案，可大幅缩短运行时间，附上部分代码

[myhhub]

更多参照：https://github.com/myhhub/InStock
下面给一些示例：

原始的：
    data.loc[:, 'ma250'] = pd.Series(tl.MA(data['close'].values, 250), index=data.index.values)
优化后的：
   data.loc[:, 'ma250'] = tl.MA(data['close'].values, timeperiod=250)

原始的：
    for i in range(1, len(data)):
        if data.iloc[i - 1]['p_change'] >= 9.5 and data.iloc[i]['p_change'] >= 9.5:
            return True
优化后的：
for _p_change in data['p_change'].values:
        if _p_change >= 9.5:
            if previous_p_change >= 9.5:
                return True
            else:
                previous_p_change = _p_change
        else:
            previous_p_change = 0.0

原始的：
for index, row in data.iterrows():
        if row['close'] > highest_row['close']:
            highest_row = row
        elif row['close'] < lowest_row['close']:
            lowest_row = row
优化后的：
    for _close, _volume, _date in zip(data['close'].values, data['volume'].values, data['date'].values):
        if _close > highest_row[0]:
            highest_row[0] = _close
            highest_row[1] = _volume
            highest_row[2] = _date
        elif _close < lowest_row[0]:
            lowest_row[0] = _close
            lowest_row[1] = _volume
            lowest_row[2] = _date

原始的：
    for i in range(1, len(data)):
        if data.iloc[i - 1]['p_change'] < -7 \
                or (data.iloc[i]['close'] - data.iloc[i]['open']) / data.iloc[i]['open'] * 100 < -7 \
                or data.iloc[i - 1]['p_change'] + data.iloc[i]['p_change'] < -10 \
                or (data.iloc[i]['close'] - data.iloc[i - 1]['open']) / data.iloc[i - 1]['open'] * 100 < -10:
优化后的：
    for _p_change, _close, _open in zip(data['p_change'].values, data['close'].values, data['open'].values):
        if _p_change < -7 or (_close - _open) / _open * 100 < -7 \
                or previous_p_change + _p_change < -10 \
                or (_close - previous_open)/previous_open * 100 < -10:
            return False

[sngyai]

这可真是太棒了！另外方便提个PR吗？