# ------------------------------------------------------------------------ #
# Copyright 2022 SPTK Working Group #
# #
# Licensed under the Apache License, Version 2.0 (the "License"); #
# you may not use this file except in compliance with the License. #
# You may obtain a copy of the License at #
# #
# http://www.apache.org/licenses/LICENSE-2.0 #
# #
# Unless required by applicable law or agreed to in writing, software #
# distributed under the License is distributed on an "AS IS" BASIS, #
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. #
# See the License for the specific language governing permissions and #
# limitations under the License. #
# ------------------------------------------------------------------------ #
from torch import nn
from vector_quantize_pytorch import ResidualVQ
[docs]
class MultiStageVectorQuantization(nn.Module):
"""See `this page <https://github.com/lucidrains/vector-quantize-pytorch>`_
for details.
Parameters
----------
order : int >= 0
Order of vector, :math:`M`.
codebook_size : int >= 1
Codebook size, :math:`K`.
n_stage : int >= 1
Number of stages (quantizers), :math:`Q`.
**kwargs : additional keyword arguments
See `this page`_ for details.
References
----------
.. [1] A. v. d. Oord et al., "Neural discrete representation learning," *Advances in
Neural Information Processing Systems*, pp. 6309-6318, 2017.
"""
def __init__(self, order, codebook_size, n_stage, **kwargs):
super().__init__()
assert 0 <= order
assert 1 <= codebook_size
assert 1 <= n_stage
self.vq = ResidualVQ(
dim=order + 1, codebook_size=codebook_size, num_quantizers=n_stage, **kwargs
).float()
@property
def codebooks(self):
return self.vq.codebooks
[docs]
def forward(self, x, codebooks=None, **kwargs):
"""Perform residual vector quantization.
Parameters
----------
x : Tensor [shape=(..., M+1)]
Input vectors.
codebooks : Tensor [shape=(Q, K, M+1)]
External codebooks. If None, use internal codebooks.
**kwargs : additional keyword arguments
See `this page`_ for details.
Returns
-------
xq : Tensor [shape=(..., M+1)]
Quantized vectors.
indices : Tensor [shape=(..., Q)]
Codebook indices.
losses : Tensor [shape=(Q,)]
Commitment losses.
Examples
--------
>>> x = diffsptk.nrand(4)
>>> x
tensor([-0.5206, 1.0048, -0.3370, 1.3364, -0.2933])
>>> msvq = diffsptk.MultiStageVectorQuantization(4, 3, 2).eval()
>>> xq, indices, _ = msvq(x)
>>> xq
tensor([-0.4561, 0.9835, -0.3787, -0.1488, -0.8025])
>>> indices
tensor([0, 2])
"""
if codebooks is not None:
cb = self.codebooks
for i, layer in enumerate(self.vq.layers):
layer._codebook.embed[:] = codebooks.view_as(cb)[i]
d = x.dim()
if d == 1:
x = x.unsqueeze(0)
xq, indices, losses = self.vq(x.float(), **kwargs)
if d == 1:
xq = xq.squeeze(0)
indices = indices.squeeze(0)
losses = losses.squeeze()
return xq, indices, losses