import pandas as pd
import matplotlib.pyplot as plt
from collections import Counter

info = [{"name": "E001", "gender": "man", "age": "34", "sales": "123", "income": 350},
        {"name": "E002", "gender": "feman", "age": "40", "sales": "114", "income": 450},
        {"name": "E003", "gender": "feman", "age": "37", "sales": "135", "income": 169},
        {"name": "E004", "gender": "man", "age": "30", "sales": "139", "income": 189},
        {"name": "E005", "gender": "feman", "age": "44", "sales": "117", "income": 183},
        {"name": "E006", "gender": "man", "age": "36", "sales": "121", "income": 80},
        {"name": "E007", "gender": "man", "age": "32", "sales": "133", "income": 166},
        {"name": "E008", "gender": "feman", "age": "26", "sales": "140", "income": 120},
        {"name": "E009", "gender": "man", "age": "32", "sales": "133", "income": 75},
        {"name": "E010", "gender": "man", "age": "36", "sales": "133", "income": 40}
        ]


# 读取数据
def get_data():
    df = pd.DataFrame(info)#DataFrame是一个以命名列方式组织的分布式数据集
    df[["age"]] = df[["age"]].astype(int)  # 数据类型转为int
    df[["sales"]] = df[["sales"]].astype(int)  # 数据类型转为int
    return df


def group_by_gender(df):
    var = df.groupby('gender').sales.sum()#groupby将元素通过函数生成相应的Key,数据就转化为Key-Value格式,之后将Key相同的元素分为一组
    fig = plt.figure()
    ax1 = fig.add_subplot(211)#2*1个网格，1个子图
    ax1.set_xlabel('Gender')  # x轴标签
    ax1.set_ylabel('Sum of Sales')  # y轴标签
    ax1.set_title('Gender wise Sum of Sales')  # 设置图标标题
    var.plot(kind='bar')
    plt.show()  # 显示


def group_by_age(df):
    age_list = [20, 30, 40, 50]
    res = pd.cut(df['age'], age_list, right=False)
    count_res = pd.value_counts(res)
    df_count_res = pd.DataFrame(count_res)
    print(df_count_res)
    plt.hist(df['age'], bins=age_list, alpha=0.7)  # age_list 根据年龄段统计
    # 显示横轴标签
    plt.xlabel("nums")
    # 显示纵轴标签
    plt.ylabel("ages")
    # 显示图标题
    plt.title("pic")
    plt.show()


def gender_count(df):
    res = df['gender'].value_counts()
    df_res = pd.DataFrame(res)
    label_list = df_res.index

    plt.axis('equal')
    plt.pie(df_res['gender'], labels=label_list,
            autopct='%1.1f%%',
            shadow=True,  # 设置阴影
            explode=[0, 0.1])  # 0 ：扇形不分离，0.1：分离0.1单位
    plt.title('gender ratio')
    plt.show()

    print(df_res)
    print(label_list)


if __name__ == '__main__':
    data = get_data()
    group_by_gender(data)
    gender_count(data)
    group_by_age(data)