熊猫排序功能无法识别列

Question

我正在尝试按州的字母顺序组织我的 df，但是当我使用 sort_values 按州排序时，没有任何反应。我相信数据的提取方式存在问题，因为我收到了一个无法识别“状态”的 KeyError。我应该使用 rename 函数而不是像我那样重命名列吗？

import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import datetime

def load_data():

    # importing datasets
    df_2017=pd.read_excel('assets/US_States_Visited_2017.xlsx', skiprows=6,skipfooter=13)
    df_2018=pd.read_excel('assets/US_States_Visited_2018.xlsx', skiprows=7,skipfooter=7)
    df_2019=pd.read_excel('assets/US_States_Visited_2019.xlsx', skiprows=6,skipfooter=8)
    
    # renaming columns
    df_2017.columns = ['2017_rank','state','2016_market_share','visitation_2016','2017_market_share','visitation_2017','volume_change']
    df_2018.columns = ['2018_rank','state','2018_market_share','visitation_2018','volume_change','2017_market_share','visitation_2017']
    df_2019.columns = ['2019_rank','state','2019_market_share','visitation_2019','volume_change','2018_market_share','visitation_2018']
    
    # stripping state names
    df_2017['state'] = df_2017['state'].str.strip()
    df_2018['state'] = df_2018['state'].str.strip()
    df_2019['state'] = df_2019['state'].str.strip()
    
    # dropping all columns except for relevent state and visitation columns
    df_2017.drop(df_2017.columns[[0,2,4,6]], axis=1,inplace=True)
    df_2018.drop(df_2018.columns[[0,2,4,5,6]], axis=1,inplace=True)
    df_2019.drop(df_2019.columns[[0,2,4,5,6]], axis=1,inplace=True) 
    
    # multiplying visitation by 1000 to get accurate value
    df_2017['visitation_2016'] = df_2017['visitation_2016']*1000
    df_2017['visitation_2017'] = df_2017['visitation_2017']*1000
    df_2018['visitation_2018'] = df_2018['visitation_2018']*1000
    df_2019['visitation_2019'] = df_2019['visitation_2019']*1000
    
    # starting output at state column
    df_2017=df_2017.set_index('state')
    df_2018=df_2018.set_index('state')
    df_2019=df_2019.set_index('state')
    
    # merging all datasets by state variable
    merged_US_states_visitation = df_2017.merge(df_2018,on='state',how= 'left').merge(df_2019,on='state',how='right')
    
    #sorting by name
    merged_US_states_visitation.sort_values(by=['state'])
    
    return merged_US_states_visitation

load_data().head(25)

Answer 1

当您的目标是索引时，您正在尝试 i) sort_values；和 ii) 您没有分配排序结果。一起去：

merged_US_states_visitation.sort_index(inplace=True)