CODEWITHSUNDEEP
pythonpandas
WangYang.Cao
WangYang.Cao

Reputation: 143

How to insert a row (list) into DataFrame ?

Guys,

I have a columns-defined DataFrame and a list like this : 

df = pd.DataFrame(columns=list('ABCDEF'))

[0.25, 1, [97, 99], array([18.,  16.,  17.]), array([ 31.,  30.,  29.]), array([ 0.])]

Actually every element in this list is a column for my expected DataFrame, I would like to have something like :

A       B         C                  D                    E             F
0.25    1   [97,99]  [18.,  16.,  17.]   [ 31.,  30.,  29.]         [ 0.]

And I could insert the list in a while loop, namely insert one row into DataFrame, e.g. in second loop I have another row like :

[0.25, 2, [132, 134], array([17.]), array([ 29., 30., 31.]), array([15., 16.])]

A       B           C                  D                    E             F
0.25    1     [97,99]  [18.,  16.,  17.]   [ 31.,  30.,  29.]         [ 0.]
0.25    2   [132,134]              [17.]   [ 29.,  30.,  31.]    [15., 16.]

But when I create the DataFrame, it always put element in array into one column :

                                0    A    B    C    D    E    F
0                            0.25  NaN  NaN  NaN  NaN  NaN  NaN
1                               1  NaN  NaN  NaN  NaN  NaN  NaN
2  [97.7123918594, 99.7123918594]  NaN  NaN  NaN  NaN  NaN  NaN
3  [17.0, 24.0, 18.0, 16.0, 17.0]  NaN  NaN  NaN  NaN  NaN  NaN
4              [31.0, 30.0, 29.0]  NaN  NaN  NaN  NaN  NaN  NaN
5                           [0.0]  NaN  NaN  NaN  NaN  NaN  NaN

Any way can get what I expected ? Really appreciate.

Upvotes: 0

Views: 899

Answers (1)

Zero
Zero

Reputation: 76917

You can use .loc

In [324]: df = pd.DataFrame(columns=list('ABCDEF'))

In [325]: l1 = [0.25, 1, [97, 99], array([18.,  16.,  17.]), array([ 31.,  30.,  29.]), array([ 0.])]

In [326]: df.loc[len(df.index)] = l1

In [327]: df
Out[327]:
      A  B         C                   D                   E      F
0  0.25  1  [97, 99]  [18.0, 16.0, 17.0]  [31.0, 30.0, 29.0]  [0.0]

However, if you have list of lists upfront, you can

pd.DataFrame([l1, l2, ..., ln], columns=list('abcdef'))

Upvotes: 2

Related Questions