Pandas学习笔记（2）Indexing, Selecting & Assigning

Pandas中两种常用索引函数loc和iloc，与python不同的是这两个 row-first, column-second。

iloc是Index-based selection，loc是Label-based selection

注：iloc使用Python stdlib索引模式，其中包含范围的第一个元素，排除最后一个元素。所以0:10将选择条目0…9。但是，loc中0:10将选择条目0…10。

1.将表中的description列从review放到desc中

desc = reviews.iloc[ : ,1]

或者：

desc = reviews.loc[ : ,'description']

2.将description列中第一个值赋到first_description中

first_description = reviews.iloc[0,1]

或者：

first_description = reviews.loc[0,'description']

3.选择第一行数据赋到first_row中

first_row = reviews.iloc[0,:]

或者：

first_row = reviews.loc[0,:]

4.将description这列中的前十个数据放到first_descriptions中

first_descriptions = reviews.iloc[0:10,1]

或者：

first_descriptions = reviews.loc[0:9,'description']  # 这里第一个索引是0:9，原因是在前面的注中

5.将行索引为1,2,3,5,8的数据赋到sample_reviews中

sample_reviews = reviews.iloc[[1,2,3,5,8],:]

或者：

sample_reviews = reviews.loc[[1,2,3,5,8],:]

6.将第0,1,10,100行的列为如图四个的数据赋到df中

df = reviews.iloc[[0,1,10,100],[0,5,6,7]]

或者：

df = reviews.loc[[0,1,10,100],['country','province','region_1','region_2']]

7.创建一个df中有country和variety中的前100行数据

df = reviews.iloc[0:100,[0,11]]

或者：

df = reviews.loc[0:99,['country','variety']]

8.创建一个italian_wines将包含印度的赋入：

italian_wines = reviews.loc[reviews.country == 'Italy']

9.创建一个top_oceania_wines将Australia，New Zealand中points大于等于95的赋入

top_oceania_wines = reviews.loc[reviews.country.isin(['Australia','New Zealand']) & (reviews.points >= 95)]

补充：

1.assign操作

reviews['critic'] = 'everyone'
reviews['critic']

0         everyone
1         everyone
            ...   
129969    everyone
129970    everyone
Name: critic, Length: 129971, dtype: object

2.数据非空，用notnull函数

reviews.loc[reviews.price.notnull()]

3.set_index()函数，更换第一列的索引

reviews.set_index("title")