我想计算一件商品出现在购物车中的频率。我有一个可以被buyerid识别的东西。此buyerid可以购买多个物品(也可以购买两次、三次、...、n次)。可由itemid和description识别。我想计算一下一件商品最终被放在购物车里的次数。例如,在5次购买中,有3人购买了苹果,即0.6%。我想把这笔钱花在所有的产品上,我该怎么做呢?
import pandas as pd
d = {'buyerid': [0,0,1,2,3,3,3,4,4,4,4],
'itemid': [0,1,2,1,1,1,2,4,5,1,1],
'description': ['Banana', 'Apple', 'Apple', 'Strawberry', 'Apple', 'Banana', 'Apple', 'Dog-Food', 'Beef', 'Banana', 'Apple'], }
df = pd.DataFrame(data=d)
display(df.head(20))

我的尝试:
# How many % of the articels are the same?
# this is wrong... :/
df_grouped = df.groupby('description').count()
display(df_grouped)
df_apple = df_grouped.iloc[0]
percentage = df_apple[0] / df.shape[0]
print(percentage)
[OUT] 0.45454545454545453数学公式
count of all buys (count_buy ) = 5
count how many an apple appears in the buy (count_apple) = 3
count_buy /count_apple = 3 / 5 = 0.6我想要什么(请注意,我没有计算这些值,这些只是一些无用的值)

发布于 2020-11-11 18:10:10
使用GroupBy.size并除以buyerid的唯一值的计数除以Series.nunique
print (df.groupby(['itemid','description']).size())
itemid description
0 Banana 1
1 Apple 3
Banana 2
Strawberry 1
2 Apple 2
4 Dog-Food 1
5 Beef 1
dtype: int64
purch = df['buyerid'].nunique()
df1 = df.groupby(['itemid','description']).size().div(purch).reset_index(name='percentage')
print (df1)
itemid description percentage
0 0 Banana 0.2
1 1 Apple 0.6
2 1 Banana 0.4
3 1 Strawberry 0.2
4 2 Apple 0.4
5 4 Dog-Food 0.2
6 5 Beef 0.2发布于 2020-11-11 18:10:47
我将对其进行分组并创建一个新列,如下所示:
df_grp = df.groupby('description')['buyerid'].sum().reset_index(name='total')
df_grp['percentage'] = (df_grp.total / df_grp.total.sum()) * 100
df_grp结果:
description total percentage
0 Apple 11 39.285714
1 Banana 7 25.000000
2 Beef 4 14.285714
3 Dog-Food 4 14.285714
4 Strawberry 2 7.142857发布于 2020-11-11 18:20:25
像往常一样,有多种方法可以获得金牌,但我将按如下方式进行旋转:您的输入:
import pandas as pd
d = {'buyerid': [0,0,1,2,3,3,3,4,4,4,4],
'itemid': [0,1,2,1,1,1,2,4,5,1,1],
'description': ['Banana', 'Apple', 'Apple', 'Strawberry', 'Apple', 'Banana', 'Apple', 'Dog-Food', 'Beef', 'Banana', 'Apple'], }
df = pd.DataFrame(data=d)在下一步中,使用buyer_id作为index,description作为列来透视数据,并将NA替换为0
df2 = df.pivot_table(values='itemid', index='buyerid', columns='description', aggfunc='count')
df2 = df2.fillna(0)导致
description Apple Banana Beef Dog-Food Strawberry
buyerid
0 1.0 1.0 0.0 0.0 0.0
1 1.0 0.0 0.0 0.0 0.0
2 0.0 0.0 0.0 0.0 1.0
3 2.0 1.0 0.0 0.0 0.0
4 1.0 1.0 1.0 1.0 0.0在桌面上调用平均值:
df_final = df2.mean()结果
description
Apple 1.0
Banana 0.6
Beef 0.2
Dog-Food 0.2
Strawberry 0.2
dtype: float64https://stackoverflow.com/questions/64784234
复制相似问题