filter by array numpy

Question

I am trying to filter my ndarray by another array I have collected (with the same values)

My main ndarray looks like

[['Name' 'Col1' 'Count']
 ['test' '' '413']
 ['erd' ' ' '60']
 ..., 
 ['Td1' 'f' '904']
 ['Td2' 'K' '953']
 ['Td3' 'r' '111']]

I have another list with various matching names

names = ['Td1','test','erd']

What I'd Like to Do

I'd like to use the list names as a filter against the ndarray above?

What I've Tried

name_filter = main_ndarray[:,0] == names

This does not work

What I'd Expect

[['Name' 'Col1' 'Count']
 ['test' '' '413']
 ['erd' ' ' '60']
 ['Td1' 'f' '904']]

Community · Accepted Answer · 2017-05-23 12:10:03Z

3

Consider using Pandas for this kind of data:

import pandas as pd

data = [['Name', 'Col1', 'Count'],
        ['test', '', '413'],
        ['erd', ' ', '60'],
        ['Td1', 'f', '904'],
        ['Td2', 'K', '953'],
        ['Td3', 'r', '111']]

df = pd.DataFrame(data[1:], columns=data[0])
names = ['Td1','test','erd']
result = df[df.Name.isin(names)]

Results:

>>> df
   Name Col1 Count
0  test        413
1   erd         60
2   Td1    f   904
3   Td2    K   953
4   Td3    r   111
>>> result
   Name Col1 Count
0  test        413
1   erd         60
2   Td1    f   904
>>>

References

edited May 23, 2017 at 12:10

CommunityBot

11 silver badge

answered Aug 3, 2015 at 14:03

YXD

32.6k15 gold badges79 silver badges117 bronze badges

Sign up to request clarification or add additional context in comments.

Comments

Nitish · Accepted Answer · 2015-08-03 13:58:45Z

1

You can use the filter function too.

cats_array = numpy.array(
 [['Name' ,'Col1', 'Count'],
 ['test', '' ,'413'],
 ['erd' ,' ' ,'60'],
 ['Td1' ,'f' ,'904'],
 ['Td2' ,'K' ,'953'],
 ['Td3' ,'r', '111']]
 )

 names = ['Td1','test','erd']

 filter(lambda x: x[0] in names, cats_array)

gives:

[array(['test', '', '413'],
       dtype='|S5'), array(['erd', ' ', '60'],
       dtype='|S5'), array(['Td1', 'f', '904'],
       dtype='|S5')]

answered Aug 3, 2015 at 13:58

Nitish

7,4561 gold badge18 silver badges15 bronze badges

2 Comments

cat Over a year ago

So out of curiosity what would I do with that array now? it didn't retain the "regular" formatting. It now has this dtype in there and the values in index 1 are completely separate from a contiguous array

Nitish Over a year ago

@cat You can hit it with map(lambda a: list(a), filter(lambda x: x[0] in names, cats_array)) to keep a list-like formatting. If you do that, your answer will be [['test', '', '413'], ['erd', ' ', '60'], ['Td1', 'f', '904']]

Cleb · Accepted Answer · 2016-06-14 08:47:23Z

1

I would also go with @YXD's Pandas solution but just for the sake of completeness I also provide a simple solution based on list comprehension:

data = [['Name', 'Col1', 'Count'],
 ['test', '', '413'],
 ['erd', ' ', '60'],
 ['Td1', 'f', '904'],
 ['Td2', 'K', '953'],
 ['Td3', 'r', '111']]

names = ['Td1', 'test', 'erd']

# select all sublist of data
res = [l for l in data if l[0] in names]

# insert the first row of data
res.insert(0, data[0])

which then gives you the desired output:

[['Name', 'Col1', 'Count'],
 ['test', '', '413'],
 ['erd', ' ', '60'],
 ['Td1', 'f', '904']]

edited Jun 14, 2016 at 8:47

answered Aug 3, 2015 at 14:41

Cleb

26.2k23 gold badges128 silver badges164 bronze badges

Collectives™ on Stack Overflow

filter by array numpy

3 Answers 3

Comments

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

3 Answers 3

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related