How to convert a pandas list row to numpy array?

Question

I have a dataframe that has two columns:

name, [sizes]

Example

one, [1,2,3]
two, [4,5,6]

I would like to convert values to array for further processing.

x_data = list(set(df.name.values))
y_data = [df.query('name == @h').sizes.values  for h in x_data]

This returns:

[array([list([1, 2, 3])], dtype=object), array([list([4,5,6])], dtype=object)]

Is there a way to flatten this out and have it as a simple array[] instead of array[list[]]?

Quang Hoang · Accepted Answer · 2020-10-29 14:11:32Z

1

Use tolist():

np.array(df['sizes'].tolist())

Output:

array([[1, 2, 3],
       [4, 5, 6]])

Note it only works if all the lists are of the same size within the column. For different sizes, we would get something like:

array([list([1, 2, 3]), list([4, 5, 6, 7])], dtype=object)

answered Oct 29, 2020 at 14:11

Quang Hoang

151k11 gold badges64 silver badges86 bronze badges

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

1 Answer 1