Implementation of AudioLM, a SOTA Language Modeling Approach to Audio Generation out of Google Research, in Pytorch
2.36k
stars
255
forks
source link
HubertWithKmeans -- attribute self.output_layer never defined in __init__ #139
Closed
tchamb closed 1 year ago
https://github.com/lucidrains/audiolm-pytorch/blob/0240f6a6bd4db34ed98799d18623ae4b5606b62a/audiolm_pytorch/hubert_kmeans.py#L81
Hi,
The output_layer parameter for hubert was added in 0.25.1 but the parameter is never saved to an attribute, causing an error during forward pass.