Vector quantization is a large area that use many algorithms and theoretical results (from Fejes Toth's hexagonal grid to Peter Gruber's quantization on manifolds, including non-uniform and non-isotropic source results, KNN, SOM, different metrics, etc).
I know that:
- in tensorflow the layer could be quantized as a tensor object,
- in pytorch there is useful methods for standart quantization.
But what if I'm interested to implemet my own algorithm, some special grids?
Should I rewrite layer in my own way, or there is interface for such tasks?
Dear Andrei Zakharov
This paper might be helpful, have a look:
https://res.mdpi.com/d_attachment/sensors/sensors-21-02984/article_deploy/sensors-21-02984-v3.pdf
Kind Regards
Qamar Ul Islam
Perhaps I may add that I always found Article Development of Adaptive Fuzzy-Neuro Generalized Learning-Vec...
helpful, as they provided a rigorous section on various test approaches including known data sets, but also specifically synthesised data for the testing. Hope this helps, Marcus
https://machinelearningmastery.com/learning-vector-quantization-for-machine-learning/
- Badges
- Science topic