为什么我的代码运行这么慢？ （嵌套循环）

Question

我最近写了一些代码，它正在创建我想要的输出，但是这需要一个永恒的……我有70万个客户线要遍历435个仓库，而运行10万个花了3个小时。

我知道这个问题可能含糊不清，但是我不确定为什么它运行如此缓慢。我怀疑这是由于我嵌套了for循环造成的，但是除非将其分块，否则数据不会加载。

import pandas as pd
import geopy.distance
dfware = pd.read_csv('dfware.csv', encoding = "ISO-8859-1")

dfcust = pd.read_csv(r'dfcust.csv', encoding = "ISO-8859-1")
ppmwinarray = []
#Chunk size to load in
csize=10 ** 3

Bigoutput = []

y=0
for dfcust in pd.read_csv(r'dfcust.csv', encoding = "ISO-8859-1", chunksize = csize):
    #For all columns (index) and rows, in datacust, iterate through those rows.
    y+=1
    print(y)
    ppmwinarray = []
    z=0
    for index,row in dfcust.iterrows():
        #Assign the Lattitude variable in the row titled 'lat' to lat1
        lat1 = row['Lat']
        # Assign the longitude variable in the row titled 'Long' to lon1
        lon1 = row['Lon']
        dlist=[]
        dindex=[]
        print(z)
        z+=1

        for index2, row2 in dfware.iterrows():
            y+=1
            lat2 = row2['Lat']
            lon2 = row2['Lon']
            coords_1 = [lat1, lon1]
            coords_2 = [lat2, lon2]
            distance = geopy.distance.distance(coords_1, coords_2).miles
            if distance > 300:
                distance = 0
            else:
                distance = distance
            dlist.append(distance)
            d_i = ((300-distance)/300)
            if d_i != 1:
                d_i=d_i
            else:
                d_i=0
                dindex.append(d_i)
            sumdi = sum(dindex)
            if sumdi == 0:
                sumdi = 1
            #Defining last 13 as PPM Index
            ppmdindex =(dindex[-13:])
            #Calculating their independent win chance
            IndependentWinChance=[ x/ sumdi for x in ppmdindex]
            #Store in an array
            ppmarray = IndependentWinChance
            #Summing independent chances to get sum chance
            sumppmWinChance = sum(ppmarray)
            #Appending the sum of all distance indexes
            ppmarray.append(sumdi)
            #Appending the sum of ppm win chance
            ppmarray.append(sumppmWinChance)
            ppmwinarray.append(ppmarray)
    Bigoutput.extend(ppmwinarray)


Bigoutputdf = pd.DataFrame(Bigoutput)
Bigoutputdf.to_csv('customers1234.csv')
exit()