How can we utilize the autoencoder for practical purposes?
- It turns out that encoded representations (embeddings) given by the encoder are magnificent objects for similarity retrieval.
- With such a condensed encoding representation, simple vector similarity measures between embeddings (such as cosine similarity) will create much more human-interpretable similarities between images.
https://github.com/ankonzoid/artificio/tree/master/image_retrieval
재밌는게 재밌는거다