Creating a numpy array which contains list of arrays for .npz input file

Question

I am doing an analysis of food expenditures by 5 different members belonging to 5 different age group. I want to create a file in .npz format, which should have two variables, viz., 'age' and 'person'. I am trying to get an array containing a list of arrays.

I created a list of 5 members 'person' and a list of 5 corresponding age group as 'age'. However, after accessing the created .npz file, I am getting a combined array of size (5,7).

person1 = np.array([(1, 2, 3, 4),
                    (4, 5, 6, 5),
                    (7, 8, 9, 6),
                    (9, 6, 5, 4),
                    (6, 5, 4, 3),
                    (6, 5, 4, 3),
                    (4, 3, 5, 7)],
                    dtype=[('BF', '<f8'), ('Lunch', '<f8'), ('Snacks', '<f8'), ('Dinner', '<f8')])
person2 = person1
person3 = person1
person4 = person1
person5 = person1

person = [person1, person2, person3, person4, person5]

age = [10, 20, 30, 40, 50]

np.savez('test.npz', age=age, person=person)

with np.load('test.npz', allow_pickle=False) as data:
    list_person = data['person']
    age_group = data['age']
    # df = pd.DataFrame(list_person)
    # df.to_excel('test.xlsx', index=True)

I am expecting 'list_person' as an array of size (5,). Each element of which should have an array of shape (7,4). So that while exporting in excel I get (5,1) data.

hpaulj · Accepted Answer · 2019-08-30 03:46:11Z

savez makes arrays of all list inputs; so that's what you'll see upon load:

In [105]: np.array(person).shape                                                                             
Out[105]: (5, 7)
In [106]: np.array(person).dtype                                                                             
Out[106]: dtype([('BF', '<f8'), ('Lunch', '<f8'), ('Snacks', '<f8'), ('Dinner', '<f8')])
In [107]: np.array(age).shape                                                                                
Out[107]: (5,)
In [108]: np.array(age).dtype                                                                                
Out[108]: dtype('int64')

person is constructed from 5 copies of person1, so the result is (5,7), and dtype has 4 fields (those aren't dimensions).

In [112]: df = pd.DataFrame(person1)                                                                         
In [113]: df                                                                                                 
Out[113]: 
    BF  Lunch  Snacks  Dinner
0  1.0    2.0     3.0     4.0
1  4.0    5.0     6.0     5.0
2  7.0    8.0     9.0     6.0
3  9.0    6.0     5.0     4.0
4  6.0    5.0     4.0     3.0
5  6.0    5.0     4.0     3.0
6  4.0    3.0     5.0     7.0

Trying to make a dataframe from the (5,7) array produces an error. Flattening it to (35,) does work.

Collectives™ on Stack Overflow

Creating a numpy array which contains list of arrays for .npz input file

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related