Pandas操作案例

numpy基础

import numpy as np

arr = np.array([1, 2, 4, 5])
print(type(arr))
print(arr)
print(np.nan)
print(np.where(arr>2))
print(arr[np.where(arr>2)[0]])

pandas读取数据

import pandas as pd

def create_dict_df():
        data = {'jobname': ['大数据开发工程师', '数据分析师', 'Python爬虫分析师', '大数据运维工程师'],
                'salary': [20000, 18000, 16000, 17000], 'city': ['北京', '北京', '上海', '郑州']}
        df = pd.DataFrame(data)
        print(type(df))
        print(df)
        df.to_excel('data.xlsx')
        df.to_csv('data.csv')

def df_view():
        path = 'salaryinfo.xlsx'
        df = pd.read_excel(path)
        df = df[['jobname','salary','city']]
        print(df)

def df_view2():
        path = 'salaryinfo_sheet.xlsx'
        # df = pd.read_excel(path,sheet_name=[0,1,2])
        # df = pd.read_excel(path,sheet_name=['北京','上海'])
        dict1 = pd.read_excel(path,sheet_name=None)
        print('***')
        print(dict1)
        df = pd.concat(dict1,ignore_index=True)
        print(df)
        print(type(df))

def df_view3():
        path = 'salaryinfo_sheet.xlsx'
        df = pd.read_excel(path,usecols=['jobname','salary','city'],
                           converters={'salary':salary_format})
        print(df)
        print(type(df))

def salary_format(salary):
        return f'{salary}元'



if __name__ == '__main__':
    # create_dict_df()
    # df_view3()
    df_view()

 DataFrame操作

import pandas as pd


def df_view():
    path = 'salaryinfo.xlsx'
    df = pd.read_excel(path)
    print(df)

    # df = df[['jobname','salary','city']]
    # df = df['jobname']
    # print(df)
    high_data = df[(df['salary'] >= 15000) & (df['salary'] < 20000)][['jobname', 'salary']]
    print(type(high_data))
    print(high_data)


def df_view1():
    path = 'salaryinfo_sheet.xlsx'
    df = pd.read_excel(path)
    # 二维结构维度信息
    print(df.shape)
    # 查看df的信息（列、空值信息、列数据类型）
    print(df.info())
    # 列数据类型
    print(df.dtypes)
    # 查看空值
    print(df.isnull())
    # 唯一值查看
    print(df['exp'].unique())
    # 查看值数据
    print(df.values)
    # 查看列名
    print(df.columns)
    # 查看前10条数据
    print(df.head(10))
    # 查看末尾5条数据
    print(df.tail())
    # 数据筛选
    print(df[df['edu'] == '本科以上'])
    print(df.loc[df['edu'] == '本科以上'])
    # 获取数据
    print(df.loc[0])
    # 遍历数据
    for index, row in df.iterrows():
        print(row)
        print('岗位名称:', row['jobname'])


if __name__ == '__main__':
    # create_dict_df()
    df_view1()

综合案例 

import pandas as pd
import matplotlib.pyplot as plt
from pyecharts import options as opts
from pyecharts.charts import Map


def df_deal():
    path = 'salaryinfo_sheet.xlsx'
    df = pd.read_excel(path, sheet_name=None)
    df = pd.concat(df, ignore_index=True)
    # print(df.head())
    # df = df.dropna()
    df['edu'] = df['edu'].fillna('未知')

    df['edu'] = df['edu'].map(str.strip)
    # print(df.head())
    # df['edu'] = df['edu'].apply(lambda x:x.strip())
    df['jobname'] = df['jobname'].apply(lambda x: x.lower())
    df = df.rename(columns={'jobname': '岗位名称'})
    # print(df.head())

    # df = df.drop_duplicates()
    # print(df['city'].drop_duplicates())

    # df.drop([])
    df.drop('company', axis=1, inplace=True)
    print(df)

    df['edu'] = df['edu'].replace('本科以上', '本科')
    df['city'] = df['city'].replace('郑州', '河南')
    print(df)
    df = df.sort_values(by=['salary'], ascending=False)
    print(df)
    df_stat = df.groupby('city')['salary'].mean()
    print(df_stat)


    plt.rcParams['font.sans-serif'] = ['simhei']
    plt.title('不同城市大数据平均薪资')
    plt.pie(df_stat.values,labels=df_stat.index)
    # plt.bar(df_stat.index,df_stat.values)
    # plt.scatter(df_stat.index, df_stat.values)
    plt.show()

    city_ = map_visualmap(df_stat.index,df_stat.values)
    city_.render(path="test_map_1.html")






def map_visualmap(provinces, value) -> Map:
    c = (
        Map()
            .add("", [list(z) for z in zip(provinces, value)], "china")
            .set_global_opts(
            title_opts=opts.TitleOpts(title="不同省份大数据平均薪资分级设色图"),
            visualmap_opts=opts.VisualMapOpts(max_=2000),
        )
    )
    return c


if __name__ == '__main__':
    # create_dict_df()
    df_deal()