pandas versus excel 学习总结3

行操作

• reset_index(drop=True)抛弃旧索引，生成新的
• append中ignore_index=True在行追加序列时设置
• drop(index=[])删除某几行
• student.iloc[]与切片student[]作用相同
• iloc:通过行号索引行数据
• loc:通过行标签索引数据
• students.Name.at[]:行号索引定位到某列的某一行

import pandas as pd

page_001 = pd.read_excel('Students.xlsx', sheet_name='Page_001')
page_002 = pd.read_excel('Students.xlsx', sheet_name='Page_002')

# 追加已有
students = page_001.append(page_002).reset_index(drop=True)

# 追加新建
stu = pd.Series({'ID': 41, 'Name': 'Abel', 'Score': 90})
students = students.append(stu, ignore_index=True)

# 删除（可切片）
students = students.drop(index=[39, 40])

# 插入
stu = pd.Series({'ID': 100, 'Name': 'Bailey', 'Score': 100})
part1 = students[:21]  # .iloc[] is the same
part2 = students[21:]
students = part1.append(stu, ignore_index=True).append(part2).reset_index(drop=True)

# 更改
stu = pd.Series({'ID': 101, 'Name': 'Danni', 'Score': 101})
students.iloc[39] = stu

# 设置空值
for i in range(5, 15):
    students['Name'].at[i] = ''

# 去掉空值
missing = students.loc[students['Name'] == '']
students.drop(missing.index, inplace=True)

print(students)

列操作

• pd.concat([’’,’’],axis=)追加
• pd.drop(index=,columns=)删除行或者列
• pd.insert(列号索引位置,column=列名,value=列值)
• dropna丢弃有空值的行

import pandas as pd
import numpy as np

page_001 = pd.read_excel('Students.xlsx', sheet_name='Page_001')
page_002 = pd.read_excel('Students.xlsx', sheet_name='Page_002')

# 追加列
students = pd.concat([page_001,page_002],axis=).reset_index(drop=True)
students['Age'] = 25
# students['Age'] = pd.Series(range(0,len(students)))

# 删除列
students.drop(columns=['Age'],inplace=True)
# 插入列
students.insert(1,column='Foo',value=range(0,len(students)))
# 改列名
# students.rename(columns={"Foo":"FOO","Name":'name'},inplace=True)
# 设置空值
students.ID = students.ID.astype(float)
for i in range(5,15):
    students.ID.at[i]=np.nan
# 去掉空值
students.dropna(inplace=True)

print(students)