Euclidean distance between features vectors

Question

I have a dataset such as: `

team  y            
A     African Dance    [[1.059685349464416, 0.328705966472625, 0.3115...
      Ballet           [[0.486603736877441, 1.678925514221191, 0.0157...
      Contemporary     [[0.06553386151790601, 2.121821165084839, 0, 0...

B     African Dance    [[1.129618763923645, 0.775617241859436, 0.0577...
      Ballet           [[1.164714455604553, 0.6662477850914, 0, 0.138...
      Contemporary     [[0.050464563071727, 0.856616079807281, 0, 0.3...

`

I wanna go through each row to calculate euclidean distance between all 2 pairs of array instances in a specific row.

`

 for i in range(features_vectors.size):
   for j in range(len(features_vectors[i])-1):
     fv1 = np.array(features_vectors[i][j])
     fv2 = np.array(features_vectors[i][j+1])
     print(np.linalg.norm(fv1 - fv2))

`

but I know in this way it won't see all the instances in an array because I want to calculate the distance between [0][0] and [0][1], then [0,0] and [0,2], and so on. how should I use the nested loop to see the data in this order?

Answer 1

This will skip duplicates:

for i,fv1 in enumerate(features_vectors):
    for fv2 in features_vectors[i+1:]:
        print(np.linalg.norm(fv1 - fv2))

You haven't really told us anything about the input data (is it one column? Several columns?), so this might need adapting.