It seems that you are dealing with big numbers, since it raised the error RuntimeWarning
. To get rid of such errors, as a numpythonic way you can first calculate the sum of each row using the np.sum()
function by specifying the proper axis then repeat and reshape the array in order to be able to divide with your array, them multiple with 100 and round the result:
col, row = np.shape(let_mat)
let_prob = np.round((let_mat/np.repeat(let_mat.sum(axis=1),row).reshape(col, row).astype(float))*100,2)
Demo :
>>> a = np.arange(20).reshape(4,5)
>>>
>>> a
array([[ 0, 1, 2, 3, 4],
[ 5, 6, 7, 8, 9],
[10, 11, 12, 13, 14],
[15, 16, 17, 18, 19]])
>>> np.round((a/np.repeat(a.sum(axis=1),5).reshape(4,5).astype(float))*100,2)
array([[ 0. , 10. , 20. , 30. , 40. ],
[ 14.29, 17.14, 20. , 22.86, 25.71],
[ 16.67, 18.33, 20. , 21.67, 23.33],
[ 17.65, 18.82, 20. , 21.18, 22.35]])