Advanced pandas processing-grouping and aggregation
Grouping and aggregation is usually a way to analyze data. It is usually used together with some statistical functions to view the grouping of data [In pandas, it must be connected together and cannot be used alone. It is meaningless to talk about grouping without aggregation]
Think about the fact that the cross-tab and pivot table just now also have the function of grouping, so they are a form of grouping, but they mainly count the number of times or calculate the ratio! ! See the effect:
1 What grouping and aggregation
2 Group API
- DataFrame.groupby(key, as_index=False) [as_index: whether to index] [data can be grouped multiple times, you need to pass a list inside to complete]
- key: grouped column data, can be multiple
- Case: Price data of different pens in different colors
col =pd.DataFrame({'color': ['white','red','green','red','green'], 'object': ['pen','pencil','pencil','ashtray','pen'],'price1':[5.56,4.20,1.30,0.56,2.75],'price2':[4.75,4.12,1.60,0.75,3.15]})
color object price1 price2
0 white pen 5.56 4.75
1 red pencil 4.20 4.12
2 green pencil 1.30 1.60
3 red ashtray 0.56 0.75
4 green pen 2.75 3.15
- Group, group colors, aggregate price
# 分组,求平均值
col.groupby(['color'])['price1'].mean()
col['price1'].groupby(col['color']).mean() # 这两种方式是一样的
color
green 2.025
red 2.380
white 5.560
Name: price1, dtype: float64
# 分组,数据的结构不变
col.groupby(['color'], as_index=False)['price1'].mean()
color price1
0 green 2.025
1 red 2.380
2 white 5.560
3 Starbucks retail store data
Now we have a set of statistics about Starbucks stores around the world. What if I want to know the number of Starbucks in the US and China, or I want to know the number of Starbucks in each province in China, then what should I do?
Data source: https://www.kaggle.com/starbucks/store-locations/data
3.1 Data acquisition
Read Starbucks store data from the file
# 导入星巴克店的数据
starbucks = pd.read_csv("./data/starbucks/directory.csv")
3.2 Perform group aggregation
# 按照国家分组,求出每个国家的星巴克零售店数量
count = starbucks.groupby(['Country']).count() # 因为此行输出一样,所以取任意一行进行统计(Brand)
Draw a picture to show the result
count['Brand'].plot(kind='bar', figsize=(20, 8))
plt.show()
Suppose we join the provinces and cities to group together
# 设置多个索引,set_index()
starbucks.groupby(['Country', 'State/Province']).count()
Take a closer look at this structure. Which structure is similar to the one mentioned earlier? ?
Similar to the previous MultiIndex structure
4 summary
- groupby for data grouping
- In pandas, it is meaningless to talk about grouping without aggregation