Append array to beginning of another array

Question

I'm attempting to perform a simple task: append an array to the beginning of another array. Here a MWE of what I mean:

a = ['a','b','c','d','e','f','g','h','i']
b = [6,4,1.,2,8,784.,43,6.,2]
c = [8,4.,32.,6,1,7,2.,9,23]
# Define arrays.
a_arr = np.array(a)
bc_arr = np.array([b, c])
# Append a_arr to beginning of bc_arr
print np.concatenate((a_arr, bc_arr), axis=1)

but I keep getting a ValueError: all the input arrays must have same number of dimensions error.

The arrays a_arr and bc_arr come like that from a different process so I can't manipulate the way they are created (ie: I can't use the a,b,c lists).

How can I generate a new array of a_arr and bc_arr so that it will look like:

array(['a','b','c','d','e','f','g','h','i'], [6,4,1.,2,8,784.,43,6.,2], [8,4.,32.,6,1,7,2.,9,23])

• Tejas Pendse: Can I ask why you're using a numpy array to store mixed datatypes, especially chars? I'm not sure most of the numpy functionality would be available to you if you use it. Why not use a simple list or a custom class?
• Gabriel: I'ts related to this question I made yesterday: stackoverflow.com/questions/23707082/…
• Tejas Pendse: I still don't understand what advantage you'd have using a numpy array...
• Padraic Cunningham: do b and c have to be joined when you create arrays?
• Gabriel: @PadraicCunningham no, each initial array (a_arr, bc_arr) should stay as a sub-array within the final array like show in the question.

Answer 1

Can you do something like.

In [88]: a = ['a','b','c','d','e','f','g','h','i']
In [89]: b = [6,4,1.,2,8,784.,43,6.,2]
In [90]: c = [8,4.,32.,6,1,7,2.,9,23]
In [91]: joined_arr=np.array([a_arr,b_arr,c_arr],dtype=object)
In [92]: joined_arr
Out[92]: 
array([['a', 'b', 'c', 'd', 'e', 'f', 'g', 'h', 'i'],
 [6.0, 4.0, 1.0, 2.0, 8.0, 784.0, 43.0, 6.0, 2.0],
 [8.0, 4.0, 32.0, 6.0, 1.0, 7.0, 2.0, 9.0, 23.0]], dtype=object)

• Gabriel: This seems to work, I can use bc_arr instead of b_arr,c_arr. Thanks.
• Padraic Cunningham: No worries. I missed the need to keep bc_arr as it is.

Answer 2

this should work

In [84]: a=np.atleast_2d(a).astype('object')
In [85]: b=np.atleast_2d(b).astype('object')
In [86]: c=np.atleast_2d(c).astype('object')
In [87]: np.vstack((a,b,c))
Out[87]:
array([[a, b, c, d, e, f, g, h, i],
 [6.0, 4.0, 1.0, 2.0, 8.0, 784.0, 43.0, 6.0, 2.0],
 [8.0, 4.0, 32.0, 6.0, 1.0, 7.0, 2.0, 9.0, 23.0]], dtype=object)