数据集操作以查找 Python 中的温度差异

Question

我有以下数据集：

oxford, 20180101, tempmax, 140
oxford, 20180101, tempmin, 40
oxford, 20180102, tempmax, 160
oxford, 20180102, tempmin, 20
oxford, 20180103, tempmax, 90
oxford, 20180103, tempmin, 10
oxford, 20180104, tempmax, 200
oxford, 20180104, tempmin, 120

我怎样才能读入这个文件，按位置和日期组合，并计算出 python 中的温度差异？

例如 output：

oxford, 20180101, tdiff, 40
oxford, 20180102, tdiff, 160
oxford, 20180103, tdiff, 80
oxford, 20180104, tdiff, 80

python 中的哪种操作可以用于此？

Answer 1

尝试这个

import pandas as pd

df = pd.DataFrame({"A":["oxford","oxford","oxford","oxford","oxford","oxford","oxford","oxford"],
                   "B":["20180101","20180101","20180102","20180102","20180103","20180103","20180104","20180104"],
                    "C":["tempmax","tempmin","tempmax","tempmin","tempmax","tempmin","tempmax","tempmin"],
                    "D":[140,40,160,20,90,10,200,120]})


df["B"] = pd.to_datetime(df["B"])
dfnew = df.sort_values(["B"]).groupby(['A', 'B'])["D"].apply(lambda x: (max(list(x))-min(list(x)))).reset_index()
dfnew["diff"] = "tdiff"


//output

        A          B    D   diff
0  oxford 2018-01-01  100  tdiff
1  oxford 2018-01-02  140  tdiff
2  oxford 2018-01-03   80  tdiff
3  oxford 2018-01-04   80  tdiff