1. 相关知识点

1.1 修改DataFrame列名

data=pd.DataFrame(student_data,columns=['student_id','age'])

1.2 获取行列数

players.shape

1.3 显示前n行

employees.head(n)

1.4 条件数据选取值

students.loc[students['student_id']==101,['name','age']]
students[students['student_id']==101][['name','age']]
students.query('`student_id`==101')[['name','age']]

1.5 创建新列

• 处理数据的时候，根据已知列得到新的列，可以考虑使用pandas.DataFrame.assign()函数
• 使用assign函数不会改变原数据，而是返回一个新的DataFrame对象，包含所有现有列和新生成的列
• 注意：assign和apply函数的主要区别在于前者不改变原数据，apply函数是在原数据的基础上添加新列

employees['bonus']=employees['salary'].apply(lambda x:x*2)
employees=employees.assign(bonus=employees.salary*2)
employees['bonus']=employees['salary']*2

1.6 删去重复的行

customers.drop_duplicates(subset=['email'],keep='first')

1.7 删除空值的数据

# axis=0代表行
students.dropna(subset=['name'],how='any', axis=0,inplace = False)

1.9 修改列名

data=data.rename(columns={'Dest':'iata_code','index':'from'})
students.columns=['student_id','first_name','last_name','age_in_years']

1.10 修改数据类型

students['grade']=students['grade'].astype('int')

1.11 填充缺失值

products['quantity'].fillna(0,inplace=True)
   # products.replace({'quantity':{
    # None:0
    # }},inplace=True)

1.12 数据上下合并

df1._append(df2)
# pd.concat([df1,df2],axis=0)

1.13 pivot_table透视表的使用

weather.pivot_table(index='month',values='temperature',columns='city',aggfunc='sum')

1.14 melt透视表的使用

• df.pivot() 将长数据集转换成宽数据集，df.melt() 则是将宽数据集变成长数据集

pd.melt(report,id_vars['product'],var_name='quarter',value_name='sales')

1.15 条件查询及排序

animals[animals['weight'] > 100].sort_values(by='weight', ascending=False)

2. 题目

2.1 从表中创建 DataFrame((Pandas 数据结构)

import pandas as pd

def createDataframe(student_data: List[List[int]]) -> pd.DataFrame: 
    data=pd.DataFrame(student_data,columns=['student_id','age'])
    return data
student_data=[[1,15],[2,11],[3,11],[4,20]]

print(createDataframe(student_data))

2.2 获取 DataFrame 的大小(数据检验)

import pandas as pd

def getDataframeSize(players: pd.DataFrame) -> List[int]:
    return list(players.shape)

2.3 显示前三行(数据检验)

import pandas as pd

def selectFirstRows(employees: pd.DataFrame) -> pd.DataFrame:
    return employees.head(3)

2.4 数据选取(数据选取)

import pandas as pd

def selectData(students: pd.DataFrame) -> pd.DataFrame:
    return students.loc[students['student_id']==101,['name','age']]
    # return students[students['student_id']==101][['name','age']]
    # return students.query('`student_id`==101')[['name','age']]

2.5 创建新列(数据选取)

import pandas as pd

def createBonusColumn(employees: pd.DataFrame) -> pd.DataFrame:
    employees['bonus']=employees['salary'].apply(lambda x:x*2)
    # employees=employees.assign(bonus=employees.salary*2)
    # employees['bonus']=employees['salary']*2
    return employees

2.6 删去重复的行(数据清理)

import pandas as pd

def dropDuplicateEmails(customers: pd.DataFrame) -> pd.DataFrame:
    return customers.drop_duplicates(subset=['email'],keep='first')

2.7 删去丢失的数据(数据清理)

import pandas as pd

def dropMissingData(students: pd.DataFrame) -> pd.DataFrame:
    # axis=0代表行
    return students.dropna(subset=['name'],how='any', axis=0,inplace = False)

2.8 修改列(数据清理)

import pandas as pd

def modifySalaryColumn(employees: pd.DataFrame) -> pd.DataFrame:
    employees=employees.assign(salary=employees.salary*2)
    # employees['salary']=employees['salary'].apply(lambda x:x*2)
    # employees['salary']=employees['salary']*2
    return employees

2.9 重命名列(数据清理)

import pandas as pd

def renameColumns(students: pd.DataFrame) -> pd.DataFrame:
    students.columns=['student_id','first_name','last_name','age_in_years']
    # dic={
    #     'id':'student_id',
    #     'first':'first_name',
    #     'last':'last_name',
    #     'age':'age_in_years'}
    # students=students.rename(columns=dic)
    return students

2.10 改变数据类型(数据清理)

import pandas as pd

def changeDatatype(students: pd.DataFrame) -> pd.DataFrame:
    students['grade']=students['grade'].astype('int')
    return students

2.11 填充缺失值(数据清理)

import pandas as pd

def fillMissingValues(products: pd.DataFrame) -> pd.DataFrame:
    products['quantity'].fillna(0,inplace=True)
    # products['quantity']=products['quantity'].fillna(0)
    # products.replace({'quantity':{
    # None:0
    # }},inplace=True)
    return products

1.12 重塑数据：连结(表格重塑)

import pandas as pd

def concatenateTables(df1: pd.DataFrame, df2: pd.DataFrame) -> pd.DataFrame:
    return df1._append(df2)
    # return pd.concat([df1,df2],axis=0)

1.13 数据重塑：透视(表格重塑)

import pandas as pd

def pivotTable(weather: pd.DataFrame) -> pd.DataFrame:
    return weather.pivot_table(index='month',values='temperature',columns='city',aggfunc='sum')
    # return weather.set_index(['month','city']).unstack()['temperature

2.14 重塑数据：融合(表格重塑)

import pandas as pd

def meltTable(report: pd.DataFrame) -> pd.DataFrame:
    report=pd.melt(report,id_vars=['product'],var_name='quarter',value_name='sales')
    return report

2.15 方法链(高级技巧)

import pandas as pd

def findHeavyAnimals(animals: pd.DataFrame) -> pd.DataFrame:
    animals = animals[animals['weight'] > 100].sort_values(by='weight', ascending=False)
    return animals[['name']]