ConvTranspose2d

class ConvTranspose2d(in_channels, out_channels, kernel_size, stride=1, padding=0, output_padding=0, dilation=1, groups=1, bias=True, conv_mode='cross_correlation', compute_mode='default', dtype=None, **kwargs)

Quantized version of ConvTranspose2d.

Applies a 2D transposed convolution over a quantized input tensor, used for inference only.

The parameter is same with ConvTranspose2d but dtype.

Parameters

dtype – data type of the output, should be qint8.

classmethod from_qat_module(qat_module)

return a QuantizedModule instance converted from a QATModule instance.

previous

ConvBnRelu2d

next

BatchMatMulActivation