ConvDecoder#

class deepinv.models.ConvDecoder(img_shape, in_size=(4, 4), layers=7, channels=256)[source]#

Bases: Module

Convolutional decoder network.

The architecture was introduced in “Accelerated MRI with Un-trained Neural Networks”, and it is well suited as a deep image prior (see deepinv.models.DeepImagePrior).

Parameters:

  • img_shape (tuple) – shape of the output image.

  • in_size (tuple) – size of the input vector.

  • layers (int) – number of layers in the network.

  • channels (int) – number of channels in the network.

forward(x, scale_out=1)[source]#

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Examples using ConvDecoder:#

Reconstructing an image using the deep image prior.

Reconstructing an image using the deep image prior.