18 lines
539 B
Python
18 lines
539 B
Python
|
# flake8: noqa: F401
|
||
|
|
r"""Quantized Modules.
|
||
|
|
|
||
|
|
This file is in the process of migration to `torch/ao/nn/quantized`, and
|
||
|
|
is kept here for compatibility while the migration process is ongoing.
|
||
|
|
If you are adding a new entry/functionality, please, add it to the
|
||
|
|
appropriate file under the `torch/ao/nn/quantized/modules`,
|
||
|
|
while adding an import statement here.
|
||
|
|
"""
|
||
|
|
|
||
|
|
from torch.ao.nn.quantized.modules.utils import (
|
||
|
|
_hide_packed_params_repr,
|
||
|
|
_ntuple_from_first,
|
||
|
|
_pair_from_first,
|
||
|
|
_quantize_weight,
|
||
|
|
WeightedQuantizedModule,
|
||
|
|
)
|