评价此页

conv2d#

class torch.ao.nn.quantized.functional.conv2d(input, weight, bias, stride=1, padding=0, dilation=1, groups=1, padding_mode='zeros', scale=1.0, zero_point=0, dtype=torch.quint8)[source]#

对量化的二维输入(由多个输入通道组成)应用二维卷积。

有关详细信息和输出形状,请参阅 Conv2d

参数
  • input – 量化输入张量,形状为 (minibatch,in_channels,iH,iW)(\text{minibatch} , \text{in\_channels} , iH , iW)

  • weight – 量化滤波器,形状为 (out_channels,in_channelsgroups,kH,kW)(\text{out\_channels} , \frac{\text{in\_channels}}{\text{groups}} , kH , kW)

  • bias – **非量化**偏置张量,形状为 (out_channels)(\text{out\_channels})。张量类型必须是 torch.float

  • stride – 卷积核的步幅。可以是单个数字或元组 (sH, sW)。默认为 1

  • padding – 输入两侧的隐式填充。可以是单个数字或元组 (padH, padW)。默认为 0

  • dilation – 核元素之间的间距。可以是单个数字或元组 (dH, dW)。默认为 1

  • groups – 将输入分成组,in_channels\text{in\_channels} 应能被组数整除。默认为 1

  • padding_mode – 要使用的填充模式。目前量化卷积仅支持“zeros”。默认值:“zeros”

  • scale – 输出的量化尺度。默认值:1.0

  • zero_point – 输出的量化零点。默认值:0

  • dtype – 要使用的量化数据类型。默认值:torch.quint8

示例

>>> from torch.ao.nn.quantized import functional as qF
>>> filters = torch.randn(8, 4, 3, 3, dtype=torch.float)
>>> inputs = torch.randn(1, 4, 5, 5, dtype=torch.float)
>>> bias = torch.randn(8, dtype=torch.float)
>>>
>>> scale, zero_point = 1.0, 0
>>> dtype_inputs = torch.quint8
>>> dtype_filters = torch.qint8
>>>
>>> q_filters = torch.quantize_per_tensor(filters, scale, zero_point, dtype_filters)
>>> q_inputs = torch.quantize_per_tensor(inputs, scale, zero_point, dtype_inputs)
>>> qF.conv2d(q_inputs, q_filters, bias, padding=1, scale=scale, zero_point=zero_point)