CODEWITHSUNDEEP
pythonpython-3.xpandaslistdataframe
Shaida Muhammad
Shaida Muhammad

Reputation: 1650

Create Pandas DataFrame from a list and list of lists

I have two python lists

messages = ['message1', 'message2', 'message3']

labels = [[1,0,1,3,1], [1,1,2,0,3], [0,0,2,1,0]]

I am creating dataFrame which will take messages as first column and labels as cat_1, cat_2, cat_3, cat_4, cat_5 i.e. total 6 columns

I tried

msgs_labels = pd.DataFrame(
{'message': messages,
 'cat': labels,
})

but it returns two columns. messages and cat.

Upvotes: 1

Views: 273

Answers (3)

jezrael
jezrael

Reputation: 862641

If no problem with starting by 0 for new columns names use DataFrame constructors with join:

df = pd.DataFrame({'message': messages}).join(pd.DataFrame(labels).add_prefix('cat_'))
print (df)
    message  cat_0  cat_1  cat_2  cat_3  cat_4
0  message1      1      0      1      3      1
1  message2      1      1      2      0      3
2  message3      0      0      2      1      0

f = lambda x: f'cat_{x + 1}'
df = pd.DataFrame({'message': messages}).join(pd.DataFrame(labels).rename(columns=f))

print (df)

    message  cat_1  cat_2  cat_3  cat_4  cat_5
0  message1      1      0      1      3      1
1  message2      1      1      2      0      3
2  message3      0      0      2      1      0

Some another ideas:

f = lambda x: f'cat_{x + 1}'
df = (pd.DataFrame(labels,index=messages)
        .rename(columns=f)
        .rename_axis('messages')
        .reset_index())

print (df)
   messages  cat_1  cat_2  cat_3  cat_4  cat_5
0  message1      1      0      1      3      1
1  message2      1      1      2      0      3
2  message3      0      0      2      1      0

Or a bit crazy:

f = lambda x: f'cat_{x + 1}'
df = (pd.DataFrame(labels,index=pd.Series(messages, name='messages'))
        .rename(columns=f)
        .reset_index())

Or solution with processing nested lists first:

d = {f'cat_{i + 1}': x for i, x in enumerate(map(list, zip(*labels)))}
d = {**{'message': messages}, **d}

df = pd.DataFrame(d)
print (df)
    message  cat_1  cat_2  cat_3  cat_4  cat_5
0  message1      1      0      1      3      1
1  message2      1      1      2      0      3
2  message3      0      0      2      1      0

Upvotes: 4

rhug123
rhug123

Reputation: 8768

Here is a way:

df = pd.DataFrame({i:j for i,j in zip(messages,labels)}).T.add_prefix('cat_').rename_axis('messages',axis=0).reset_index()

Upvotes: 0

Anurag Dabas
Anurag Dabas

Reputation: 24314

Use:

df=(pd.DataFrame(labels,index=messages)
   .reset_index()
   .rename(columns=lambda x:'cat_'+str(x+1) if x!='index' else 'messages'))

Output of df:

    messages    cat_1   cat_2   cat_3   cat_4   cat_5
0   message1    1       0       1       3       1
1   message2    1       1       2       0       3
2   message3    0       0       2       1       0

Upvotes: 1

Related Questions