【数据分析&数据挖掘】pandas交叉表&透视表

一、交叉表

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)

# 构建交叉表
# index ——指定某列进行行分组
# columns ——指定某列进行列分组
# 如果不传其他参数，只传index 与columns, 那么交叉表构建 index 与 columns的相对数量

res = pd.crosstab(
    index=detail.loc[:, "dishes_name"],
    columns=detail.loc[:, "amounts"],
)
print("res: 
", res)

# 以index这一列为行分组，以columns这一列为列分组，对values进行统计aggfunc指标
# values 与 aggfunc必须同时存在
# 只能按照单列进行行分组，只能按照按照单列进行列分组，只能统计某一列的aggfunc指标
res = pd.crosstab(
    index=detail.loc[:, "dishes_name"],
    columns=detail.loc[:, "amounts"],
    values=detail.loc[:, "counts"],
    aggfunc=np.max
)

print("res: 
", res)

二、透视表

import pandas as pd
import numpy as np

# 创建一个dataframe
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]
    },
)

res = df.groupby(by="cls_id")[["score", "height"]].mean()
print("groupby-res: 
", res)

# data ——df
# index 指定行分组 ——可以按照多列进行行分组
# values ——aggfunc 所针对的主体，可以是多列
# aggfunc ——默认为均值

# 创建透视表
res = pd.pivot_table(
    data=df,
    # index="cls_id",  # 按照行或列分组
    # columns="cls_id",
    columns=["cls_id", "group_id"],
    values=["score", "height"],
    aggfunc="mean",
    margins=True,
    margins_name="xixi"
)

print("res: 
", res)

print("~"*60)

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

res = pd.pivot_table(
    data=detail,
    # index="order_id",
    # index="cost",
    columns="dishes_id",
    # values="amounts",
    values="cost",
    aggfunc="max",
    fill_value=-1  # 将全为空的数据进行填充
)

print("res: 
", res)