pandas :如何在稀疏表中选择具有非零值的列 [英] Pandas: How to select columns with non-zero value in a sparse table

问题描述

我有一个2000列的稀疏表，我想选择一些特定的行及其非零值.我该怎么办?

I have a sparse table with 2000 column and I want to select few specific rows and their non-zero values. how can I do it?

    id       name   e1   e2   e3   e4 . . e550 . . e1200 .   e1760. . e2000
    1      engine1   0   0    0    2322   1300               140       0
    2      engine2   0   0  1230   0        0        2100              0
    .
    .
    .
    998000


df[df.name==engine2 & ? ]
    id    name      e3       e1200
    2   engine2    1230       2100   

和

df[df.name==engine1 & ? ]

    id    name      e4       e550     e1760
    1   engine1    2322       1300      140

推荐答案

您可以过滤两次-首先通过掩码m1行，然后通过m2列. 上次使用 loc :

You can filter twice - first rows by mask m1 and then column by m2. Last use loc:

m1 = (df['name'] == 'engine2')
m2 = (df[m1] != 0).all()

print (df.loc[m1,m2])
   id     name    e3  e1200
1   2  engine2  1230   2100

这篇关于 pandas :如何在稀疏表中选择具有非零值的列的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！