递归地将def函数应用于数据帧

Question

我试图“递归地”计算Cat的列值

每个循环都应计算x组的Cat列最大值（Catz）。如果日期范围变为<；=60，则Cat列值应更新为Catz+=1。我对这个过程有一点印象。一、 但是，在外部有成千上万的其他数据集不需要转换为arcpy友好格式。我对熊猫不太熟悉。你知道吗

引用了[1]：Calculate DataFrame values recursively和[2]：python pandas- apply function with two arguments to columns。我还没有完全理解Series/Dataframe的概念以及如何应用这两种结果

import pandas as pd
import numpy as np
from datetime import datetime
from datetime import datetime as dt
from datetime import timedelta
import time
from datetime import date
dict = {'x':["ASPELBJNMI", "JUNRNEXCRG", "ASPELBJNMI", "JUNRNEXCRG"], 
        'start': ["6/27/2018", "8/4/2018", "8/22/2018", "8/12/2018"], 
        'finish':["8/11/2018", "10/3/2018", "8/31/2018", "10/26/2018"],
        'DateRange':[0,0,0,0],
        'Cat':[-1,-1,-1,-1],
        'ID':[1,2,3,4]} 

df = pd.DataFrame(dict)

df.set_index('ID')
def classd(houp):
    Catz = houp.Cat.min()
    Catz +=1

    houp  = houp.groupby('x')
    for x, houp2 in houp:


        houp.DateRange  = (pd.to_datetime(houp.finish.loc[:]).min()- houp.start.loc[:]).astype('timedelta64[D]')

    houp.Cat = np.where(houp.DateRange<=60, Catz , -1)
    return houp

df['Cat'] =  df[['x','DateRange','Cat']].apply(classd, axis=1).Cat
print df

当我运行我的代码时，我得到了以下回溯

Catz=小时计() AttributeError:（“'long'对象没有属性'min'，u'出现在索引0'）

期望的结果

   OBJECTID_1 * Conc *  ID  start   finish  DateRange   Cat
1   ASPELBJNMI  LAPMT   6/27/2018   8/11/2018   45  0
2   ASPELBJNMI  KLKIY   8/22/2018   8/31/2018   9   1
15  JUNRNEXCRG  CGCHK   8/4/2018    10/3/2018   60  1
16  JUNRNEXCRG  IQYGJ   8/12/2018   10/26/2018  83  -1