How to Quantize Vectors using Kmeans?

Question

I have a bunch of entities, with each instance having 40 features, so a 40-dimensional object.

I cluster them using K-means.

Now, I need to quantize them.

I want to ask two questions:

1. How to quantize the vectors, now that I have a Kmeans model?

2. What is the importance of cluster size here.

Answer 1

You map each vector to the nearest center. These are your quants.

Quantity is ignored. A quantity could be used a single time, or even 0 with bad starting conditions.