1.5.4

the quantized embeddings should be selected after the EMA step, so th…

1.5.3

in cosine sim, update with the l2norm of the smoothed embed_avg over …

1.5.2

cleanup cosine sim

1.5.1

fix an issue with multi-headed codebooks and reduction in cluster siz…

1.5.0

for cosine sim, keep track of the exponential moving averages of the …

1.4.1

make sure get_codes_from_indices work for grouped rvq

1.4.0

add grouped residual vq technique from hifi-codec paper

1.2.3

address issue https://github.com/lucidrains/vector-quantize-pytorch/i…

1.2.2

just do an assert to make sure indices are correct for rvq cross entr…

1.2.1

some cleanup using unpack from einops