英文:
Creating Subtables from Dataframe Based on Column Value
问题
我想要创建基于"BUY"和"SELL"之间距离的动态长度的子表。在匹配这两个数据点并创建表格后，它会移动到SELL后的下一个BUY实例。
即，如果在子表中存在多个"BUY"，它将跳过这些以找到一个"SELL"点。
目标是让它看起来像这样：
英文:
I have a large dataframe that looks like this (continuing onward):
I want to create subtables that are of a dynamic length based on the distance between BUY and SELL. After it matches these two data points and creates the table it would move to the next instance of BUY after the SELL.
i.e. if multiple BUY's exist inside the subtable, it would skip over these to find a SELL point.
The goal is for it to look something like this:
答案1
得分: 1
以下是代码的翻译部分：
import pandas as pd
import numpy as np
df = pd.DataFrame({'Current Time': ['18:08:20']*16,
'Price': np.arange(157.01, 157.17, 0.01),
'Moving Avg': [np.nan]*3 + list(np.arange(157.01, 157.13, 0.01)),
'Action': [np.nan]*3 + ['BUY']*3 + ['SELL']*3 + ['BUY'] + ['SELL'] + ['BUY']*4 + ['SELL']})
# 创建具有虚假价格和移动平均值的模拟数据帧
df_new = df[df['Action'].notna()]
grp = (df_new['Action'] == 'SELL').cumsum().shift().bfill()
dd = dict(tuple(df_new.groupby(grp)))
list_of_dfs = [g for _, g in dd.items() if len(g) > 1]
list_of_dfs
输出：
[ Current Time Price Moving Avg Action
3 18:08:20 157.04 157.01 BUY
4 18:08:20 157.05 157.02 BUY
5 18:08:20 157.06 157.03 BUY
6 18:08:20 157.07 157.04 SELL,
Current Time Price Moving Avg Action
9 18:08:20 157.10 157.07 BUY
10 18:08:20 157.11 157.08 SELL,
Current Time Price Moving Avg Action
11 18:08:20 157.12 157.09 BUY
12 18:08:20 157.13 157.10 BUY
13 18:08:20 157.14 157.11 BUY
14 18:08:20 157.15 157.12 BUY
15 18:08:20 157.16 157.13 SELL]
英文:
Try this:
import pandas as pd
import numpy as np
df = pd.DataFrame({'Current Time': ['18:08:20']*16,
'Price':np.arange(157.01,157.17,.01),
'Moving Avg':[np.nan]*3+list(np.arange(157.01,157.13,.01)),
'Action':[np.nan]*3+['BUY']*3+['SELL']*3+['BUY']+['SELL']+['BUY']*4+['SELL']})
#built mock dataframe with fake prices and moving aver
df_new = df[df['Action'].notna()]
grp = (df_new['Action'] == 'SELL').cumsum().shift().bfill()
dd = dict(tuple(df_new.groupby(grp)))
list_of_dfs = [g for _, g in dd.items() if len(g) > 1]
list_of_dfs
Output:
[ Current Time Price Moving Avg Action
3 18:08:20 157.04 157.01 BUY
4 18:08:20 157.05 157.02 BUY
5 18:08:20 157.06 157.03 BUY
6 18:08:20 157.07 157.04 SELL,
Current Time Price Moving Avg Action
9 18:08:20 157.10 157.07 BUY
10 18:08:20 157.11 157.08 SELL,
Current Time Price Moving Avg Action
11 18:08:20 157.12 157.09 BUY
12 18:08:20 157.13 157.10 BUY
13 18:08:20 157.14 157.11 BUY
14 18:08:20 157.15 157.12 BUY
15 18:08:20 157.16 157.13 SELL]
通过集体智慧和协作来改善编程学习和解决问题的方式。致力于成为全球开发者共同参与的知识库，让每个人都能够通过互相帮助和分享经验来进步。
评论