Odtnhj

Suppose that we have some data that contain features for the sample numbers 0,1,2,...,9 and a list y that contains the labels of 0,1,...,9 that correspond to each raw of data.These data have dimensions (2006,256) where 256 are the features and 2006 the samples numbers.

Also we are given the means of 0,1,2,...,9 as v0,v1,v2,...,v9, where each v0,v1,...,v9 has dimension (1,256).

I would like to calculate the euclidean difference between all the 2006 sample and v0,v1,v2,...,v9 in order to make classification.

In order to do that I have to take the difference between features and v0,v1,v2,...,v9 for each sample.For example for sample 1 I have to take the difference of 256 features with v0, then with v1 etc. and then find the min difference.

I define ListV=[v0,v1,v2,...,v9]
and my code is

diff=
ListV=[v0,v1,v2,v3,v4,v5,v6,v7,v8,v9]
for j in range(0,10):
 i=np.where(y==j)
 for k in range(0,len(i[0][:])):
 for l in range(0,9-j):
 diff.append(distance.euclidean(ListV[j+l],data[i[0][k],:]))

but this code leaves out many euclidean differences.

I would like some help in order to fix that and find all the differences.

This problem occurred for me once. On that case I had null data in my data-set, be sure your data have defined and real values.