I do not quite understand what this index_values object is. But if it was ndarray or could be converted to ndarray , you can simply do this:
>>> s = numpy.arange(20) >>> index_values = (numpy.random.random((3, 3)) * 20).astype('i') >>> s[index_values] = 4 >>> s array([ 0, 1, 4, 4, 4, 5, 6, 4, 8, 4, 4, 11, 12, 13, 4, 15, 4, 4, 4, 19])
Edit: But it looks like it won't work in this case. Based on your changes and comments, here is a method that I think might work for you. Random list of lists with different lengths ...
>>> index_values = [list(range(x, x + random.randrange(1, 5))) ... for x in [random.randrange(0,50) for y in range(33)]]
... not hard to convert to an array:
>>> index_value_array = numpy.fromiter(itertools.chain(*index_values), dtype='i')
If you know the length of the array, specify count for better performance:
>>> index_value_array = numpy.fromiter(itertools.chain(*index_values), dtype='i', count=83)
Since your editing indicates that you want histogram-like behavior, simple indexing will not be performed, as Robert Kern pointed out. So use numpy.histogram :
>>> hist = numpy.histogram(index_value_array, bins=range(0, 51))
histogram really built for floating point histograms. This means that the bit should be slightly larger than expected, because the last value is included in the last bit, and therefore 48 and 49 will be in the same box if we use a more intuitive range(0, 50) . The result is a tuple with an array of n samples and an array of n + 1 hopper boundaries:
>>> hist (array([2, 2, 1, 2, 1, 0, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1, 1, 5, 5, 5, 3, 3, 3, 3, 3, 2, 1, 0, 2, 3, 3, 1, 0, 2, 3, 2, 2, 2, 3, 2, 1, 1, 2, 2, 2, 0, 0, 0, 1, 0]), array([ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50]))
Now we can scale the counters 4 times and add a vector:
>>> s = numpy.arange(50, dtype='f') >>> hist[0] * 4.1 + s array([ 8.2, 9.2, 6.1, 11.2, 8.1, 5. , 6. , 7. , 12.1, 13.1, 14.1, 15.1, 16.1, 13. , 18.1, 19.1, 20.1, 37.5, 38.5, 39.5, 32.3, 33.3, 34.3, 35.3, 36.3, 33.2, 30.1, 27. , 36.2, 41.3, 42.3, 35.1, 32. , 41.2, 46.3, 43.2, 44.2, 45.2, 50.3, 47.2, 44.1, 45.1, 50.2, 51.2, 52.2, 45. , 46. , 47. , 52.1, 49. ])
I have no idea if this is suitable for your purposes, but it seems like a good approach and probably should happen at close speed c, as it uses only numpy and itertools .