CODEWITHSUNDEEP
pythonlistpandasdataframe
Liza
Liza

Reputation: 971

How to work with python list of lists?

I have a following sample list of lists:

In [] : list1
Out [] : 
[[1.0],
 [2.1],
 [3.3, 5.5, 0.69],
 [0.69, 0.9]]

I want to extract only the sublists where number of elements are equal and greater than 2 and want to store them in a data frame.

So, I expect a df something like below:

In [] : df
Out [] : 
        seq_no       items
            1        3.3 , 5.5, 0.69
            2        0.69, 0.9

Tried: 

item for item in list1 where(len(item) >2)

shows error.

Please let me know if anything is unclear.

Upvotes: 2

Views: 188

Answers (4)

Zero
Zero

Reputation: 76917

In [755]: df = pd.DataFrame({'items': [x for x in list1 if len(x)>=2]})

In [756]: df
Out[756]:
              items
0  [3.3, 5.5, 0.69]
1       [0.69, 0.9]

Add, seq_no

In [759]: df['seq_no'] = df.index + 1

In [760]: df
Out[760]:
              items  seq_no
0  [3.3, 5.5, 0.69]       1
1       [0.69, 0.9]       2

If you need a string of comma separated items

In [769]: pd.DataFrame({'items': [', '.join(map(str, x)) for x in list1 if len(x)>=2]})
Out[769]:
            items
0  3.3, 5.5, 0.69
1       0.69, 0.9

Upvotes: 9

jezrael
jezrael

Reputation: 862511

Use list comprehension and for seq column range:

a = [x for x in L if len(x) >=2]
df = pd.DataFrame({'seq':range(1, len(a)+1), 'items':a}, columns=['seq','items'])
print (df)
   seq                items
0    1     [3.3, 5.5, 0.69]
1    2     [0.69, 0.9]

Upvotes: 6

piRSquared
piRSquared

Reputation: 294218

You can store it in a pd.Series first, then filter and transform.

s = pd.Series(list1)
pd.DataFrame(s[s.str.len().ge(2)].tolist())

      0     1     2
0  0.00  0.50  0.69
1  0.69  0.88  1.00
2  1.00  1.10   NaN
3  1.10  2.00   NaN
4  2.00  2.50  2.90

And to join them

s = pd.Series(list1)
s[s.str.len().ge(2)].apply(lambda x: ', '.join(map(str, x)))

2      0.0, 0.5, 0.69
3     0.69, 0.88, 1.0
4            1.0, 1.1
8            1.1, 2.0
10      2.0, 2.5, 2.9
dtype: object

Upvotes: 5

perigon
perigon

Reputation: 2095

This should do it (with a list comprehension):

import pandas as pd
df = pd.Series([sublist for sublist in list1 if len(sublist) >= 2])

Then you can add

df.index += 1

to adjust the start index if you'd like.

Upvotes: 4

Related Questions