第 1 部分：系列和数据框

如何安装

pip install pandas

如何使用图书馆

import pandas as pd

系列

• 具有相同数据类型元素的一维数组

数据框

• 每列具有不同数据类型元素的二维数组
• 面向列

创建一个系列

1.直接地

ser = pd.Series([10, 20, 30, 40])type(ser)>> ser.index>> RangeIndex(start=0, stop=4, step=1)

• pd.Series() : 将列表放入系列

2. 标签系列索引

grades = ['A', 'B+', 'A+', 'A']ser = pd.Series(grades, index=['sam', 'june', 'max', 'july'])print(ser[0], ser[2]) # access with int indexprint(ser['sam'], ser['max']) # access with label>> A A+ser.index>> Index(['sam', 'june', 'max', 'july'], dtype='object')

3.字典到系列

grades = {'sam':'A', 'june':'B+', 'max':'A+', 'july':'A'}ser = pd.Series(grades)

创建一个数据框

1.无列标签

import pandas as pdimport numpy as npdf = pd.DataFrame(np.zeros((4,3)))type(df)>> 

2.带列标签

import pandas as pdimport numpy as npclasses = ['science', 'english', 'math']df = pd.DataFrame(np.zeros((4,3)), columns = classes)

3. 使用字典和列表

import pandas as pdnames = ['sam', 'june', 'max', 'july']korean = [100, 95, 90, 85]english = [85, 90, 95, 100]math = [90, 85, 100, 95]df = pd.DataFrame({'name':names, 'korean':korean, 'english':english, 'math':math})

4. 标签的行索引

import pandas as pdnames = ['sam', 'june', 'max', 'july']korean = [100, 95, 90, 85]english = [85, 90, 95, 100]math = [90, 85, 100, 95]df = pd.DataFrame({'korean':korean, 'english':english, 'math':math}, index=names)

访问数据框

1.系列的列索引

# continuedname_ser = df['name']type(name_ser)>> 

2. DataFrame 的列索引

# continuedclasses_df = df[['name', 'math']]type(classes_df)>> 

3.广播

# continueddf['english'] = df['english'] + 10

4. 获得一排

# continuedprint(df[1:2])print(df[2:])

5. 获取元素

# Continued# Column - Row# Elementprint(df['korean'][1])# Row - Column# Seriesprint(df[1:2]['korean'])

6. 布尔索引

# continued (4) Row Index to Label bool_index = df['math'] > 90print(bool_index)print(df[bool_index])

bool_index = df > 90print(bool_index)print(df[bool_index])

DataFrame 高级索引

• 面向行
• loc : 标签索引
• iloc : int 索引

df.loc[row index]df.loc[rwo index, column index]df.loc[row slicing]df.loc[boolean list]

forloc 高级索引示例

# continued (4) Row Index to Labeldf.loc['sam']

df.loc[['sam', 'max']]df.loc['sam', 'korean']

df.loc['june':'july'] # includes july! df.loc[df['korean']>90]

df.loc[['sam', 'june'],['math','english']]df.loc[df['math']>90,['korean','english']]

iloc 高级索引的示例

# continued (4) Row Index to Labeldf.iloc[0]df.iloc[[0,2]]df.iloc[0,0]

df.iloc[0:2] # does NOT include max! df.iloc[0:2, 1:3]df.iloc[[0,3],[1,2]]

df.iloc[0]>90df.iloc[0][df.iloc[0]>90]# sam's subject over 90 

关注七爪网，获取更多APP/小程序/网站源码资源！