gnes.encoder.numeric.vlad module

class gnes.encoder.numeric.vlad.VladEncoder(num_clusters, using_faiss_pred=False, *args, **kwargs)[source]

Bases: gnes.encoder.base.BaseNumericEncoder

batch_size = 2048
encode(vecs, *args, **kwargs)[source]
Return type:ndarray
faiss_index()[source]
kmeans_pred(vecs)[source]
kmeans_train(vecs)[source]
train(vecs, *args, **kwargs)[source]

Train the model, need to be overrided