Skip to main content
 首页 » 编程设计

python之过滤值的访问索引

2024年12月31日7TianFang

在这里,我根据谓词访问系列的值:

import numpy as np 
 
s = pd.Series(np.array([1,2,3])) 
print(type(s)) 
 
print([i for i in s if i > 2]) 

返回:

<class 'pandas.core.series.Series'> 
[3] 

如何访问与谓词匹配的值的索引,因此在本例中索引为 2。

请您参考如下方法:

使用boolean indexing使用索引:

print (s.index[s > 2]) 
#alternative 
#print (s[s > 2].index) 
Int64Index([2], dtype='int64') 

您的解决方案应更改为 Series.items :

print([k for k, v in s.items() if v > 2]) 
[2] 

性能在更大的数据中是不同的:

np.random.seed(45) 
s = pd.Series(np.random.randint(5, size=100000)) 
 
#print (s) 
 
In [93]: %timeit ([k for k, v in s.items() if v > 2]) 
13.3 ms ± 768 µs per loop (mean ± std. dev. of 7 runs, 100 loops each) 
 
In [94]: %timeit (s.index[s > 2]) 
930 µs ± 12.2 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) 
 
In [95]: %timeit (s[s > 2].index) 
1.74 ms ± 26.1 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) 
 
In [96]: %timeit (s.index[s.values > 2]) 
742 µs ± 79.6 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) 
 
In [97]: %timeit (s.index.values[s.values > 2]) 
647 µs ± 12.6 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each) 
 
In [98]: %timeit (s[s.values > 2].index.values) 
1.51 ms ± 12.4 µs per loop (mean ± std. dev. of 7 runs, 1000 loops each)