【数据分析&数据挖掘】pandas分组聚合

import pandas as pd
import numpy as np

# 加载数据
detail = pd.read_excel("../day05/meal_order_detail.xlsx")
print("detail: 
", detail)
print("detail的列名称: 
", detail.columns)

# 删除法
# 先进行判断
drop_list = []
for column in detail.columns:
    # print(column)
    # 统计每一列非空数据的数量
    res = detail.loc[:, column].count()
    # print("res: 
", res)
    if res == 0:
        drop_list.append(column)

print(drop_list)

# 再进行删除:
detail.drop(labels=drop_list, axis=1, inplace=True)
print("删除全部为空列之后的结果: 
" ,detail.shape)
print("删除全部为空列之后的结果的列名称: 
" ,detail.columns)
print("^"*60)

# 分组进行统计指标
# 按照单列进行分组——统计菜品id的最大值
res_ = detail.groupby(by="order_id")["dishes_id"].max()
res_ = detail.groupby(by=detail["order_id"])["dishes_id"].max()

print("res_: 
", res_)

# 统计所欲python班级各个小组的平均成绩
df = pd.DataFrame(
    data={
        "cls_id": ["A", "B", "C", "A", "B", "C", "A", "B", "C", "A", "B","C"],
        "group_id": [1, 1, 1, 1, 1, 1, 2, 2, 2, 2, 2, 2],
        "name": ["xixi", "haha", "taotao", "huihui", "ranran", "Island", "Tree" ,"bamao", "simao", "hanhan", "qimao", "sanmao"],
        "score": [92, 93, 39, 89, 90.5, 80, 91, 92, 65, 73, 34.5, 56],
        "height": [165, 166, 167, 168, 152, 193, 192, 190, 173, 172, 170, 169]
    },
)
print("df: 
", df)

# 按照班级分组，统计班级的平均分
# 按照单列进行分组
res = df.groupby(by="cls_id")["score"].mean()
print(res)

# 先按照班级分组，再统计各小组的平均成绩
res = df.groupby(by=["cls_id", "group_id"])["score"].mean()
print("res: 
", res)

# 按照多列分组，既要统计成绩的平均值，又要统计身高的平均值
res = df.groupby(by=["cls_id", "group_id"])[["score", "height"]].mean()
print("res: 
", res)

# 对成绩求最大值，身高求平均值
# res = detail.agg({"counts": np.max, "height": np.mean})
# print("res: 
", res)

# 对不同的列求取不同的指标
res = detail.agg({"counts": np.sum, "amounts": np.mean})

# 对不同的列求取多个相同的指标
res = detail[["counts", "amounts"]].agg([np.max, np.mean])
#
# # 对不同单列求取不同个数的指标
res = detail.agg({"counts": [np.mean, np.max], "amounts": np.min})

print("res :
", res)

# 对某列进行指定的运算
res = detail[["counts", "amounts"]].apply(lambda x: x+1)
res = detail[["counts", "amounts"]].transform(lambda x: x+1)
# res = detail[["counts", "amounts"]].apply(lambda x, y: x+y)  # 错误的， 不能跨列运算

print('detail[["counts", "amounts"]]: 
', detail[["counts", "amounts"]])
print("res :
", res)
print(detail["counts"])