Commit 2a251506 authored by Phil Wang's avatar Phil Wang
Browse files

residual vq ready for distributed training

parent 8910dbfe

audiolm_pytorch/audiolm_pytorch.py

+1 −2
Original line number Diff line number Diff line
@@ -245,8 +245,7 @@ class SoundStream(nn.Module): 
            num_quantizers = rq_num_quantizers,

            codebook_size = codebook_size,

            kmeans_init = True,

            threshold_ema_dead_code = 2,

            sync_kmeans = False

            threshold_ema_dead_code = 2

        )



        decoder_blocks = []

setup.py

+2 −2
Original line number Diff line number Diff line
@@ -3,7 +3,7 @@ from setuptools import setup, find_packages 
setup(

  name = 'audiolm-pytorch',

  packages = find_packages(exclude=[]),

  version = '0.0.2',

  version = '0.0.3',

  license='MIT',

  description = 'AudioLM - Language Modeling Approach to Audio Generation from Google Research - Pytorch',

  author = 'Phil Wang',
@@ -21,7 +21,7 @@ setup( 
    'einops>=0.4',

    'ema-pytorch',

    'torch>=1.6',

    'vector-quantize-pytorch>=0.10.1'

    'vector-quantize-pytorch>=0.10.2'

  ],

  classifiers=[

    'Development Status :: 4 - Beta',