根據“ID”列返回上一行值以查找當前行與上一行之間的差異

Question

我是 python 的新手。 我在嘗試根據 python dataframe 中的“ID”和“時間戳”查找前一行值時遇到問題。 誰能告訴我應該如何得出解決方案？

下面是一個簡單的例子：

例如：

id: 時間戳: sapStock laststockdate laststockvalue

167777 14/12/2021 184 13/12/2021 143 169406 14/12/2021 56 13/12/2021 60

#Import Libraries
import pandas as pd

#Read CSV
df = pd.read_csv(r'C:\Users\User\OneDrive\Desktop\Test.csv')

#Preview dataframe
df

#Get last value of sapStock based on "id" and "timestamp"
df['sapStockDifference'] = df.sapStock.diff(periods=1)
df

Answer 1

我希望這可能會有所幫助：

import pandas as pd
import numpy as np

df = pd.DataFrame({'id': ['169653', '167777', '169406', '165253', '169653', '167777', '169406', '165253', '169653', '167777', '169406', '165253', '169653', '167777', '169406', '165253'],
                   'sapStock': [234, 162, 36, 47, 264, 158, 60, 46, 279, 143, 60, 46, 364, 184, 56, 46],
                   'timestamp': ['11/12/2021', '11/12/2021', '11/12/2021', '11/12/2021', '12/12/2021', '12/12/2021', '12/12/2021', '12/12/2021', '13/12/2021', '13/12/2021', '13/12/2021', '13/12/2021', '14/12/2021', '14/12/2021', '14/12/2021', '14/12/2021']})
df['timestamp'] = pd.to_datetime(df['timestamp'])
df['laststockvalue'] = df.sort_values(['timestamp']).groupby(df.id)['sapStock'].apply(lambda x: x.shift(1)).fillna(value=np.nan)
df['Difference'] = df.sort_values(['timestamp']).groupby(df.id)['sapStock'].apply(lambda x: x - x.shift(1)).fillna(value=np.nan)
print(df)

#         id  sapStock  timestamp  laststockvalue  Difference
# 0   169653       234 2021-11-12             NaN         NaN
# 1   167777       162 2021-11-12             NaN         NaN
# 2   169406        36 2021-11-12             NaN         NaN
# 3   165253        47 2021-11-12             NaN         NaN
# 4   169653       264 2021-12-12           234.0        30.0
# 5   167777       158 2021-12-12           162.0        -4.0
# 6   169406        60 2021-12-12            36.0        24.0
# 7   165253        46 2021-12-12            47.0        -1.0
# 8   169653       279 2021-12-13           264.0        15.0
# 9   167777       143 2021-12-13           158.0       -15.0
# 10  169406        60 2021-12-13            60.0         0.0
# 11  165253        46 2021-12-13            46.0         0.0
# 12  169653       364 2021-12-14           279.0        85.0
# 13  167777       184 2021-12-14           143.0        41.0
# 14  169406        56 2021-12-14            60.0        -4.0
# 15  165253        46 2021-12-14            46.0         0.0