quantized

torch_to_nnef.op.quantized

PyTorch quantized::* operators translation.

Quantized layers and primitives

Maybe usefull when looking at X

packed_params._method_names()

add

add(**kwargs)

Map PyTorch: 'quantized:add' to NNEF.

add_relu

add_relu(g, node, name_to_tensor, null_ref, **kwargs)

Map PyTorch: 'quantized:add_relu' to NNEF.

conv1d

conv1d(g, node, name_to_tensor, null_ref, inference_target, **kwargs)

Map PyTorch: 'quantized:conv1d' to NNEF.

conv1d_relu

conv1d_relu(g, node, name_to_tensor, inference_target, null_ref, **kwargs)

Map PyTorch: 'quantized:conv1d_relu' to NNEF.

conv2d

conv2d(g, node, name_to_tensor, null_ref, inference_target, **kwargs)

Map PyTorch: 'quantized:conv2d' to NNEF.

conv2d_relu

conv2d_relu(g, node, name_to_tensor, null_ref, inference_target, **kwargs)

Map PyTorch: 'quantized:conv2d_relu' to NNEF.

div

div(**kwargs)

Map PyTorch: 'quantized:div' to NNEF.

linear

linear(g, node, name_to_tensor, inference_target, **kwargs)

Map PyTorch: 'quantized:linear' to NNEF.

linear_relu

linear_relu(g, node, name_to_tensor, inference_target, **kwargs)

Map PyTorch: 'quantized:linear_relu' to NNEF.

mul

mul(**kwargs)

Map PyTorch: 'quantized:mul' to NNEF.