Pandas 之入门

教程链接：https://blog.csdn.net/toto1297488504/category_9731964.html

一、十分钟入门

1.读取文件txt，转换成数组

with open("D:/PyProjects/demo/com/njbdqn/pandas/city.txt",'r',encoding='gbk') as f:
   # text 转换成数组
   str = f.readline()
   a = str.strip("[]").replace("'","").split(",")

读取文件csv，分组统计

str = pd.read_csv("D:/PyProjects/demo/com/njbdqn/pandas/jobs_csv.csv")
    # 分组统计
 print(str['location'].value_counts())

北京 48
北京-海淀区 9
北京-朝阳区 3

2.用数组组装Series

# Series(数组) => 下标是 0-1 的一维数组
citySeries = pd.Series(a)

3.组装DF，不写index，则默认0，1，2，3 ...

# DataFrame(Series,columns,index) => 二维数组，index必须是index
citytab = pd.DataFrame(citySeries, columns=list('C'))

4.用DF获得Series

# Return a Series containing counts of unique rows in the DataFrame.
print(citytab.value_counts()) # value出现的counts

c = citySeries.value_counts(ascending=True) # 排序

5.Series的mean

 print(c.mean())

6.Series的前5个切片

print(citySeries.value_counts()[:5])