FM / modules

/usr/local/lib64/python3.6/site-packages/torch/nn/intrinsic/quantized/modules/

/usr/local/lib64/python3.6/site-packages/torch/nn/intrinsic/quantized/modules

Name	Size	Mode	Actions
__pycache__/	-	0755	rm
bn_relu.py	2320	0644	edit dl rm
conv_relu.py	5771	0644	edit dl rm
linear_relu.py	1048	0644	edit dl rm
__init__.py	253	0644	edit dl rm

Edit: /usr/local/lib64/python3.6/site-packages/torch/nn/intrinsic/quantized/modules/linear_relu.py (1048B)

import torch
import torch.nn.quantized as nnq
import torch.nn.intrinsic as nni

class LinearReLU(nnq.Linear):
    r"""
    A LinearReLU module fused from Linear and ReLU modules

We adopt the same interface as :class:`torch.nn.quantized.Linear`.

Attributes:
        Same as torch.nn.quantized.Linear

Examples::

>>> m = nn.intrinsic.LinearReLU(20, 30)
        >>> input = torch.randn(128, 20)
        >>> output = m(input)
        >>> print(output.size())
        torch.Size([128, 30])
    """
    _FLOAT_MODULE = nni.LinearReLU

def __init__(self, in_features, out_features, bias=True, dtype=torch.qint8):
        super().__init__(in_features, out_features, bias, dtype)

def forward(self, x: torch.Tensor) -> torch.Tensor:
        return torch.ops.quantized.linear_relu(
            x, self._packed_params._packed_params, self.scale, self.zero_point)

def _get_name(self):
        return 'QuantizedLinearReLU'

@classmethod
    def from_float(cls, mod):
        return super(LinearReLU, cls).from_float(mod)

import torch import torch.nn.quantized as nnq import torch.nn.intrinsic as nni class LinearReLU(nnq.Linear): r""" A LinearReLU module fused from Linear and ReLU modules We adopt the same interface as :class:`torch.nn.quantized.Linear`. Attributes: Same as torch.nn.quantized.Linear Examples:: >>> m = nn.intrinsic.LinearReLU(20, 30) >>> input = torch.randn(128, 20) >>> output = m(input) >>> print(output.size()) torch.Size([128, 30]) """ _FLOAT_MODULE = nni.LinearReLU def __init__(self, in_features, out_features, bias=True, dtype=torch.qint8): super().__init__(in_features, out_features, bias, dtype) def forward(self, x: torch.Tensor) -> torch.Tensor: return torch.ops.quantized.linear_relu( x, self._packed_params._packed_params, self.scale, self.zero_point) def _get_name(self): return 'QuantizedLinearReLU' @classmethod def from_float(cls, mod): return super(LinearReLU, cls).from_float(mod)