How to filter rows in Pandas based on specific value in 1 column

Question

I have a csv, like this (no headers):

a1   b1  3
a2   b2  5 
a3   b3  8

I want to get all rows, where values in last column is >4. How can I do that?

P.S. This is why this is not duplicated question - in the link above columns are named, my columns are unnamed.

@jezrael explain please how can I use that answer in my case? I have already seen it but did not get how to use it in my case - in that case columns are named, my columns are unnamed. — Chiefir, Jul 12 '18 at 09:20

score 1 · Accepted Answer · answered Jul 12 '18 at 09:21

1

You can use boolean indexing with iloc for select last column:

df = df[df.iloc[:, -1] > 4]
print (df)
    0   1  2
1  a2  b2  5
2  a3  b3  8

Detail:

print (df.iloc[:, -1])
0    3
1    5
2    8
Name: 2, dtype: int64

answered Jul 12 '18 at 09:21

jezrael

I have already tried something smilar, but I get this error - `TypeError: '>' not supported between instances of 'str' and 'int'` – Chiefir Jul 12 '18 at 09:27
But wait now. If he pass header=False then it will be df[2]. I think OP might have missed that. There are no dfs without columns. – Anton vBR Jul 12 '18 at 09:27
@Chiefir - Waht return `df.iloc[:, -1]` ? there are numbers? If yes, how working `df.iloc[:, -1].astype(int)` ? – jezrael Jul 12 '18 at 09:29
@Chiefir - Because maybe numbers saved as strings. – jezrael Jul 12 '18 at 09:31
1

@jezrael you were right - the column was a string, not int. – Chiefir Jul 16 '18 at 07:18

1 Answers1