Creating Prototype Vector for comparison

Question

I am trying to implement an approach following a paper which compares the content vectors of words to a prototype vector, which is representative of the entire class/cluster/type/etc. In the first step, a prototype vector is calculated and I do not quite understand how the way to acquire prototype vectors.

I referred to here to the discussion of this question: However, this post seems to answer what the prototype vector is theoretically, while I need to find a practical solution to implement.

Is there an implementation in Python // Sci-kit learn that can realize the function of

A. defining/indicate a priori or induce from training instances a prototype vector B. then input feature vectors to be compared for similarity against the prototype vector from A.?

Thank you in advance for you help.

Answer 1

I think you are looking for the Nearest Centroid Classifier: http://scikit-learn.org/dev/modules/neighbors.html#nearest-centroid-classifier