Python numpy: Convert string in to numpy array

Question

I have following String that I have put together:

v1fColor = '2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,0,0,0,3,6,0,0,0,0,0,0,0,0,0,7,5,0,0,0,2,0,0,0,0,0,0,0,0,0,0,4,3,0,0,0,1,1,0,0,0,0,0,0,0,0,0,6,6,0,0,0,2,3'

I am treating it as a vector: Long story short its a forecolor of an image histogram:

I have the following lambda function to calculate cosine similarity of two images, So I tried to convert this is to numpy.array but I failed:

Here is my lambda function

import numpy as NP
import numpy.linalg as LA
cx = lambda a, b : round(NP.inner(a, b)/(LA.norm(a)*LA.norm(b)), 3)

So I tried the following to convert this string as a numpy array:

v1fColor = NP.array([float(v1fColor)], dtype=NP.uint8)

But I ended up getting following error:

    v1fColor = NP.array([float(v1fColor)], dtype=NP.uint8)
ValueError: invalid literal for float(): 2,4,14,5,0,0,0,0,0,0,0,0,0,0,12,4,0,0,0,0,0,0,0,0,0,0,0,0,0,0,15,6,0,0,0,0,1,0,0,0,0,0,0,0,0,0,20,9,0,0,0,2,2,0,0,0,0,0,0,0,0,0,13,6,0,0,0,1,0,0,0,0,0,0,0,0,0,0,10,8,0,0,0,1,2,0,0,0,0,0,0,0,0,0,17,17,

Answer 1

You have to split the string by its commas first:

NP.array(v1fColor.split(","), dtype=NP.uint8)

Answer 2

You can do this without using python string methods -- try numpy.fromstring:

>>> numpy.fromstring(v1fColor, dtype='uint8', sep=',')
array([ 2,  4, 14,  5,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0, 12,  4,  0,
        0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0,  0, 15,  6,  0,  0,
        0,  0,  1,  0,  0,  0,  0,  0,  0,  0,  0,  0, 20,  9,  0,  0,  0,
        2,  2,  0,  0,  0,  0,  0,  0,  0,  0,  0, 13,  6,  0,  0,  0,  1,
        0,  0,  0,  0,  0,  0,  0,  0,  0,  0, 10,  8,  0,  0,  0,  1,  2,
        0,  0,  0,  0,  0,  0,  0,  0,  0, 17, 17,  0,  0,  0,  3,  6,  0,
        0,  0,  0,  0,  0,  0,  0,  0,  7,  5,  0,  0,  0,  2,  0,  0,  0,
        0,  0,  0,  0,  0,  0,  0,  4,  3,  0,  0,  0,  1,  1,  0,  0,  0,
        0,  0,  0,  0,  0,  0,  6,  6,  0,  0,  0,  2,  3], dtype=uint8)

Answer 3

You can do this:

lst = v1fColor.split(',')  #create a list of strings, splitting on the commas.
v1fColor = NP.array( lst, dtype=NP.uint8 ) #numpy converts the strings.  Nifty!

or more concisely:

v1fColor = NP.array( v1fColor.split(','), dtype=NP.uint8 )

Note that it is a little more customary to do:

import numpy as np

compared to import numpy as NP

EDIT

Just today I learned about the function numpy.fromstring which could also be used to solve this problem:

NP.fromstring( "1,2,3" , sep="," , dtype=NP.uint8 )