pandas数据帧作为json文件使用索引值

Question

我有一个由groupby创建的pandas数据帧，如下所示：airbus_df.groupby(['mfr'](.apply(lambda行：行.to_json(orient='records'((.to_frame((输出看起来像：

0
mfr 
AIRBUS  [{"mfr mdl code":"3940005","icao":"A00C7A","se...
AIRBUS CANADA LTD PTNRSP    [{"mfr mdl code":"1400010","icao":"A062EC","se...
AIRBUS HELICOPTERS DEUTSCHLAND  [{"mfr mdl code":"5620040","icao":"A32422","se...
AIRBUS HELICOPTERS INC  [{"mfr mdl code":"1145005","icao":"A1F846","se...
AIRBUS INDUSTRIE    [{"mfr mdl code":"3930402","icao":"A009A4","se...
AIRBUS SAS  [{"mfr mdl code":"3940312","icao":"A3AD89","se

我想为每个"mfr"创建一个json文件。如何使用索引名称保存为json。

df_t.index:
Index(['AIRBUS','AIRBUS CANADA LTD PTNRSP'...])

Answer 1

您可以按系列循环，并通过索引值中的键将json写入文件：

import json
for k,v in airbus_df.groupby('mfr').apply(lambda row:row.to_json(orient='records')).items():
with open(f"{k}.json", 'w') as f: 
json.dump(v, f)