转载他人的，本来自己想写，算了
import pandas as pd
df_tmp = pd.DataFrame([
{"a":"data1", "cnt":100},{"a":"data2", "cnt":200},
])
df_tmp
a cnt
data1 100
data2 200
方法一：使用apply 的参数result_type 来处理
def formatrow(row):
a = row["a"] + str(row["cnt"])
b = str(row["cnt"]) + row["a"]
return a, b
df_tmp[["fomat1", "format2"]] = df_tmp.apply(formatrow, axis=1, result_type="expand")
df_tmp
a cnt fomat1 format2
data1 100 data1100 100data1
data2 200 data2200 200data2
方法一：使用zip打包返回结果来处理
df_tmp["fomat1-1"], df_tmp["format2-2"] = zip(*df_tmp.apply(formatrow, axis=1))
df_tmp
a cnt fomat1 format2 fomat1-1 format2-2
data1 100 data1100 100data1 data1100 100data1
data2 200 data2200 200data2 data2200 200data2