python數(shù)據(jù)分析之員工個人信息可視化

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}
        ]


# 讀取數(shù)據(jù)
def get_data():
    df = pd.DataFrame(info)#DataFrame是一個以命名列方式組織的分布式數(shù)據(jù)集
    df[["age"]] = df[["age"]].astype(int)  # 數(shù)據(jù)類型轉(zhuǎn)為int
    df[["sales"]] = df[["sales"]].astype(int)  # 數(shù)據(jù)類型轉(zhuǎn)為int
    return df


def group_by_gender(df):
    var = df.groupby('gender').sales.sum()#groupby將元素通過函數(shù)生成相應(yīng)的Key,數(shù)據(jù)就轉(zhuǎn)化為Key-Value格式,之后將Key相同的元素分為一組
    fig = plt.figure()
    ax1 = fig.add_subplot(211)#2*1個網(wǎng)格，1個子圖
    ax1.set_xlabel('Gender')  # x軸標(biāo)簽
    ax1.set_ylabel('Sum of Sales')  # y軸標(biāo)簽
    ax1.set_title('Gender wise Sum of Sales')  # 設(shè)置圖標(biāo)標(biāo)題
    var.plot(kind='bar')
    plt.show()  # 顯示


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 根據(jù)年齡段統(tǒng)計
    # 顯示橫軸標(biāo)簽
    plt.xlabel("nums")
    # 顯示縱軸標(biāo)簽
    plt.ylabel("ages")
    # 顯示圖標(biāo)題
    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,  # 設(shè)置陰影
            explode=[0, 0.1])  # 0 ：扇形不分離，0.1：分離0.1單位
    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)