时间戳

首页 > 解决方案 > 调整 heikin ashi 代码错误：列表喜欢到 .loc 或 [] 与任何缺少的标签不再支持

python - 调整 heikin ashi 代码错误：列表喜欢到 .loc 或 [] 与任何缺少的标签不再支持

问题描述

我正在尝试修改我在网上找到的一些用于 heikin ashi 蜡烛的代码

代码是：

def heikin_ashi(self, df):
    heikin_ashi_df = pd.DataFrame(index=df.index.values, columns=['HE_Open', 'HE_High', 'HE_Low', 'HE_Close'])

    heikin_ashi_df['HE_Close'] = (df['Open'] + df['High'] + df['Low'] + df['Close']) / 4

    for i in range(len(df)):
        if i == 0:
            heikin_ashi_df.iat[0, 0] = df['Open'].iloc[0]
        else:
            heikin_ashi_df.iat[i, 0] = (heikin_ashi_df.iat[i - 1, 0] + heikin_ashi_df.iat[i - 1, 3]) / 2

    heikin_ashi_df['HE_High'] = heikin_ashi_df.loc[:, ['HE_Open', 'HE_Close']].join(df['High']).max(axis=1)

    heikin_ashi_df['HE_Low'] = heikin_ashi_df.loc[:, ['HE_Open', 'HE_Close']].join(df['Low']).min(axis=1)

    return heikin_ashi_df

但是这个方法返回一个新的熊猫表，我只想将新行添加到我现有的表中

这是我的尝试：

def heikin_ashi_full(self, df):

    df.append(pd.Series(name='HE_Open', dtype=float))
    df.append(pd.Series(name='HE_High', dtype=float))
    df.append(pd.Series(name='HE_Low', dtype=float))
    df.append(pd.Series(name='HE_Close', dtype=float))
    df['HE_Close'] = (df['Open'] + df['High'] + df['Low'] + df['Close']) / 4

    for i in range(len(df)):
        if i == 0:
            df.iat[0, 0] = df['Open'].iloc[0]
        else:
            df.iat[i, 0] = (df.iat[i - 1, 0] + df.iat[i - 1, 3]) / 2

    df['HE_High'] = df.loc[:, ['HE_Open', 'HE_Close', 'High']].max(axis=1)

    df['HE_Low'] = df.loc[:, ['HE_Open', 'HE_Close', 'Low']].min(axis=1)

    return df

不幸的是，该方法在该方法的第 15 行崩溃

df['HE_High'] = df.loc[:, ['HE_Open', 'HE_Close', 'High']].max(axis=1)

回溯显示在这里：

追溯

主表是从具有如下数据的文件中创建的：

Date,Open,High,Low,Close,Volume,OpenInterest
2020-08-25 06:00,15.12,15.25,14.82,15.05,3824776,0.0
2020-08-26 06:00,15.09,15.44,15.02,15.07,5933882,0.0
2020-08-27 06:00,15.22,15.33,14.672,14.84,5728962,0.0
2020-08-28 06:00,15.01,15.18,14.8605,15.12,5992532,0.0
2020-08-31 06:00,15.23,15.23,14.62,14.62,7000994,0.0
2020-09-01 06:00,14.64,15.18,14.24,15.08,7347598,0.0

有趣的是，我从互联网上获得的第一种方法似乎工作得很好而且很准确。我不熟悉熊猫，不知道我是怎么做错的，也找不到解决办法。非常感谢您的帮助，并提前感谢您的宝贵时间。

标签： pythonpandas

解决方案

我尝试了您的代码，我的第一个问题是我正在尝试添加列df.append()，但我认为这是因为没有添加列。如果我在将开盘价、最高价和最低价添加为空后尝试运行您的代码，将显示结果。数据框的列结构发生了一些变化，因此我对其进行了相应的修改。另外，我认为关于 SO 上的“heikin-ashi”的这个答案会有所帮助。

import pandas as pd
import numpy as np
import io

data = '''
Date,Open,High,Low,Close,Volume,OpenInterest
2020-08-25 06:00,15.12,15.25,14.82,15.05,3824776,0.0
2020-08-26 06:00,15.09,15.44,15.02,15.07,5933882,0.0
2020-08-27 06:00,15.22,15.33,14.672,14.84,5728962,0.0
2020-08-28 06:00,15.01,15.18,14.8605,15.12,5992532,0.0
2020-08-31 06:00,15.23,15.23,14.62,14.62,7000994,0.0
2020-09-01 06:00,14.64,15.18,14.24,15.08,7347598,0.0
'''

df = pd.read_csv(io.StringIO(data), sep=',')

o = pd.Series([], index=[], name='HE_Open', dtype=float)
h = pd.Series([], index=[], name='HE_High', dtype=float)
l = pd.Series([], index=[], name='HE_Low', dtype=float)
df = pd.concat([df, o, h, l],axis=1)
df['HE_Close'] = (df['Open'] + df['High'] + df['Low'] + df['Close']) / 4

for i in range(len(df)):
    if i == 0:
        df.iat[0, 7] = df['Open'].iloc[0]
    else:
        df.iat[i, 7] = (df.iat[i - 1, 1] + df.iat[i - 1, 4]) / 2

df['HE_High'] = df.loc[:, ['HE_Open', 'HE_Close', 'High']].max(axis=1)
df['HE_Low'] = df.loc[:, ['HE_Open', 'HE_Close', 'Low']].min(axis=1)

df
    Date    Open    High    Low     Close   Volume  OpenInterest    HE_Open     HE_High     HE_Low  HE_Close
0   2020-08-25 06:00    15.12   15.25   14.8200     15.05   3824776     0.0     15.120  15.25   14.8200     15.060000
1   2020-08-26 06:00    15.09   15.44   15.0200     15.07   5933882     0.0     15.085  15.44   15.0200     15.155000
2   2020-08-27 06:00    15.22   15.33   14.6720     14.84   5728962     0.0     15.080  15.33   14.6720     15.015500
3   2020-08-28 06:00    15.01   15.18   14.8605     15.12   5992532     0.0     15.030  15.18   14.8605     15.042625
4   2020-08-31 06:00    15.23   15.23   14.6200     14.62   7000994     0.0     15.065  15.23   14.6200     14.925000
5   2020-09-01 06:00    14.64   15.18   14.2400     15.08   7347598     0.0     14.925  15.18   14.2400     14.785000

推荐阅读