泰迪杯赛前培训之Pandas

可以理解成一个字典

查看是否安装

!pip list

导入包

import pandas as pd

import numpy as np

生成随机数

pd.Series(np.random.rand(5))

查看类型

s1 = pd.Series(np.random.rand(5))
type(s1)

 在jupyter上进行

!pip list

import pandas as pd 
import numpy as np

#  方法1：通过数组创建Series
s1 = pd.Series(np.random.rand(5))
# print(s1)
print(s1,type(s1))
print('-'*50)
print(type(np.random.rand(5)))

# 方法2：通过dict创建Series
dic = {"a":1,"b":2,"c":3,"5":2}
# print(dic)
s2 = pd.Series(dic)
print(s2,type(s2))
# dtype是自行推断

#  方法3:通过标量
s3 = pd.Series(10,index=["a","c","d"])
print(s3)

# Series的属性
s = pd.Series(dic,name="series-name")
print(type(s))
print("-"*50)
# name属性
print(s.name)
print(s.rename("河北邯郸").name)  #重命名
print("-"*50)
# index属性
print(s.index,type(s.index))
print("-"*50)
# value属性
print(s.values,type(s.values))

# 位置索引
dic = {"a":1,"b":2,"c":3,"5":2}
# print(dic)
s = pd.Series(dic,name="series-name")
print(s)
print(s[1],type(s[1]),s[1].dtype)
print("-"*50)
# 修改类型
print(float(s[1]),type(float(s[1])))
print("-"*50)
# 修改指定位置
s[1] = 20
print(s)


# 切片索引
s = pd.Series(np.random.rand(5))
print(s)
print("-"*50)
print(s[1:4])
print("-"*50)
print(s[:-1])
print("-"*50)
# 指定索引的步长
print(s[::2])
# 修改值
s[:-2] = 20
print(s)

# 布尔索引
np.random.seed(88)#确保值固定
s = pd.Series(np.random.rand(5)*100)
print(s)
s[4] = None
bol = s > 55
print(bol[type(bol)])
print("-"*50)
# 通过布尔series获取值
print(s[bol])
print("-"*50)
# 查看series的方法
bol1 = s.isnull()
bol2 = s.notnull()
print(bol1)
print(bol2)
# 获取空值
print(s[bol2])

# 数据查看
np.random.seed(88)
s = pd.Series(np.random.rand(10)*100)
print(s)
print("-"*50)
print(s.head(1))
print("-"*50)
print(s.tail(2))

# 重新索引
np.random.seed(88)
s = pd.Series(np.random.rand(5)*100)
print(s)
print("-"*50)
#重新索引是修改顺序，如果不存在，就为NAN 
s1 = s.reindex(["c","d","a","e","f"])
print(s1)
print("-"*50)
s2 = s.reindex([0,"d",1,"e","f"])
print(s2)

# 对齐 
np.random.seed(88)
s1= pd.Series(np.random.rand(3)*100,index=["a","b","c"])
s2= pd.Series(np.random.rand(3)*100,index=["d","e","f"])
print(s1)
print("-"*50)
print(s2)
print("-"*50)
print(s1+s2)
print("-"*50)
s1= pd.Series(np.random.rand(3)*100,index=["a","b","c"])
s2= pd.Series(np.random.rand(3)*100,index=["a","e","f"])
print(s1 + s2)

# 删除
np.random.seed(88)
s = pd.Series(np.random.rand(5)*100,index=['a','b','c','c','c'])
print(s)
print("-"*50)
# 删除指定行
s1 = s.drop('c')
print(s1)
print("-"*50)
# 删除多行
s2 = s.drop(['a','c'])
print(s2)
print("-"*50)
s3 = s.drop('c',inplace = True)
print(s3)
print("-"*50)

# 添加
np.random.seed(88)
s1 = pd.Series(np.random.rand(3)*100,index=['a','b','c'])
s2 = pd.Series(np.random.rand(3)*100,index=['d','e','f'])
print(s1)
print("-"*50)
print(s1)
print("-"*50)
s = s1.append(s2)
print(s)

# 修改
np.random.seed(88)
s = pd.Series(np.random.rand(3)*100,index=['a','b','c'])
s['a'] = 100
s['b','c'] = 200
print(s)

Dataframe数据结构

# Dataframe是一个表格型的数据结构，“带有标签的二维数组”
# Dataframe带有index（行标签）和columns(列标签)
data = {'name':['Jack','Tom','Marry'],
        'age':[18,19,20],
        'gender':['m','m','w']}
frame = pd.DataFrame(data)
print(frame)
print(type(frame))
print(frame.index,'\n该数据类型为：',type(frame.index))
print(frame.columns,'\n该数据类型为：',type(frame.columns))
print(frame.values,'\n该数据类型为：',type(frame.values))
frame

# Dataframe创建方法1 ： 由数组list组成的字典
# 创建方法：pandas.Dataframe()
data1 = {
    'a':[1,2,3],
    'b':[3,4,5],
    'c':[5,6,7]
}
data2 = {"one":np.random.rand(3),
         "two":np.random.rand(3)}
print(data)
print("-"*50)
print(data1)
print("-"*50)
print(data2)
df1 = pd.DataFrame(data1)
print("-"*50)
print(df1)
print("-"*50)
df2 = pd.DataFrame(data2,index=['a','b','c'])
print(df2)

# Pandas数据结构Dataframe：索引
# 选择行和列
np.random.seed(88)
data = {"one":np.random.rand(3)*100,
         "two":np.random.rand(3)*100}
df = pd.DataFrame(data,index=['a','b','c'])
print(df)
print("-"*50)
# 获取列
print(df['one'],type(df['one']))
print("-"*50)
print(df[['one','two']])
# help(df)
print("-"*50)
# 获取行
print(df.loc['a'],type(df.loc['a']))
print("-"*50)
print(df.loc[['a','b']])
print("-"*50)
# 获取指定单元值
print(df.loc[['b']][['two']])

# 通过iloc获取指定行，并指定顺序
print(df.iloc[[2,1,0]])
# 切片获取行
print(df.iloc[::2])
# 布尔索引
print(df)
print(df[df>70])

# 转置
np.random.seed(88)
data = {"col1":np.random.rand(3)*100,
         "col2":np.random.rand(3)*100}
df = pd.DataFrame(data,index=['a','b','c'],columns=['col1','col2'])
print(df.T)
print("-"*50)
# 查看头尾
print(df.head(1))
print(df.tail(1))
print("-"*50)
# 添加修改
df['col3'] = 200
df['col4'] = 100
print(df)
print("-"*50)
# 删除
del df['col4']
print(df)
print("-"*50)
df1 = df.drop(['col3'],axis=1)
print(df1)