当前位置：首页 > news >正文

【Python】Pandas

news 2025/7/20 8:11:30

要求：

Pandas是基于Numpy开发出的,专门用于数据分析的开源Python库

Series(一维数据)

DataFrame(多特征数据,既有行索引,又有列索引)

DataFrame的属性

Panda数据读取(以csv为例)

pandas.read_csv(filepath_or_buffer, sep=",", names=None, usecols = None)

filepath_or_buffer : 文件路径(本地路径或url路径) sep: 分隔符 names: 列索引的名字 usecols: 指定读取的列名返回的类型: DataFrame

Dataframe通过布尔索引过滤数据

代码实现：

1、通过numpy数组创建

import pandas as pd 
import numpy as np
print(pd.Series(np.arange(4,10)))

2、通过python数组创建

import pandas as pd
import numpy as npprint(pd.Series([11,12,14],index = ["北京","上海","深圳"]))

3、通过python字典创建

import pandas as pd
import numpy as npprint(pd.Series({"北京":11,"上海":12,"深圳":14}))

4、允许索引重复

import pandas as pd
import numpy as npa = pd.Series([11,12,14],index=["北京","上海","上海"])
a["上海"]
print(a)

5、

DataFrame(多特征数据,既有行索引,又有列索引)

import pandas as pd
import numpy as npdata_3_4 = pd.DataFrame(np.arange(10,22).reshape(3,4))
print(data_3_4)
print(data_3_4[:1])
print(data_3_4[:][0])

6、DataFrame的属性

import pandas as pd
import numpy as npresult = pd.read_csv("C:/工作簿.csv",encoding='gbk')
print(result.shape)
print(result.dtypes)
print(result.ndim)
print(result.index)
print(result.columns)
print(result.values)print("-->前5个:")
print(result.head(5))
print("-->后5个:")
print(result.tail(5))
print("-->描述信息：")
print(result.describe())

7、Panda数据读取(以csv为例)

import pandas as pd
import numpy as npresult = pd.read_csv("C:/工作簿.csv",encoding='gbk')
print(result)
print(result["姓名"][0:6])
print("读取后返回的类型为-->",type(result))

8、Dataframe通过布尔索引过滤数据

import pandas as pd 
import numpy as npresult = pd.read_csv("C:/工作簿.csv",encoding='gbk')
print(result[result["age"]>23])

9、

import pandas as pd
import numpy as npIMDB_1000 = pd.read_csv("C:/IMDB-Movie-Data.csv")
#print(IMDB_1000.dtypes)
#print(IMDB_1000.sort_values(by = "Rating",ascending=False))
#print(IMDB_1000[IMDB_1000["Runtime (Minutes)"]==IMDB_1000["Runtime (Minutes)"].max()])
#print(IMDB_1000[IMDB_1000["Runtime (Minutes)"]==IMDB_1000["Runtime (Minutes)"].min()])
#print(IMDB_1000["Runtime (Minutes)"].mean())
IMDB_1000.dropna()
IMDB_1000.fillna("NaN")
IMDB_1000["Revenue (Millions)"].fillna(IMDB_1000["Revenue (Millions)"].mean(),inplace=True)
IMDB_1000