我们可以再添加一列“filestatus”，并使用 pandas 在状态列上添加逻辑吗

Question

这是我的Python代码：在下面的输出中samplefile1已传入所有 3 行，因此 newcolumn{"filestatus" : "passed"}为"InputFile":"samplefile1"因为example它有一个pass又一个fail如此新的{"filestatus":"failed"}栏目{"inputfile":"example"}import jsondf = pd.DataFrame([            ['samlefile1','[e-mail]', 'xyz' ,'pass'],            ['samlefile1','[e-mail]', 'xyz' ,'pass'],            ['samlefile1','[e-mail]', 'xyz' ,'pass'],            ['testfile','[e-mail]', 'abc' ,'pass'],            ['example','[e-mail]', 'xyz' ,'pass'],            ['example','[e-mail]', 'xyz' ,'fail']],columns = ['InputFile','UserId', 'UserGroup' ,'status']    )#Count no of status per filenameInput_status_count = df.groupby(['InputFile']).agg(success_count=('status', lambda x: x[x == 'pass'].count()),                                                           fail_count=('status', lambda x: x[x == 'fail'].count()))        #Merge the 2 files on 'InputFile' columnFinalDF = pd.merge(df, Input_status_count,on = "InputFile" )JSON_String = FinalDF.to_json(orient='records')JSON_Stringoutput:[    {"InputFile":"samlefile1","UserId":"[e-mail]","UserGroup":"xyz","status":"pass","success_count":3,"fail_count":0, "filestatus":"passed"},    {"InputFile":"samlefile1","UserId":"[e-mail]","UserGroup":"xyz","status":"pass","success_count":3,"fail_count":0, "filestaus":"passed"},    {"InputFile":"samlefile1","UserId":"[e-mail]","UserGroup":"xyz","status":"pass","success_count":3,"fail_count":0, "filestatus":"passed"},    {"InputFile":"testfile","UserId":"[e-mail]","UserGroup":"abc","status":"","success_count":1,"fail_count":0, "filestatus":"not ran"},    {"InputFile":"example","UserId":"[e-mail]","UserGroup":"xyz","status":"pass","success_count":1,"fail_count":1,"filestatus":"failed"},    {"InputFile":"example","UserId":"[e-mail]","UserGroup":"xyz","status":"fail","success_count":1,"fail_count":1, "filestatus": "failed"}]

Answer 1

是的，你可以这样做：

import numpy as np

#Count no of status per filename

Input_status_count = df.groupby(['InputFile']).agg([('filestatus', lambda x: 'passed' if x[x['status'] == 'fail'].count() == 0 else 'failed')])

Lambda 函数基本上检查我们是否获得状态为失败的零行计数，然后文件状态将被传递，否则失败。

Answer 2

您可以使用 pd.crosstab （而不是复杂的 agg）将通过和失败制成表格，并且当没有失败时状态列将通过：

Input_status_count = pd.crosstab(df['InputFile'],df['status']).reset_index()

Input_status_count['filestatus'] = ["passed" if i==0 else "failed" for i in Input_status_count['fail']]

Input_status_count

status   InputFile  fail  pass filestatus

0          example     1     1     failed

1       samlefile1     0     3     passed

2         testfile     0     1     passed