elektronn3.models.fcn_2d module

adopted from https://github.com/pochih/FCN-pytorch/blob/master/python/fcn.py LICENSE https://github.com/meetshah1995/pytorch-semseg/blob/master/LICENSE

class elektronn3.models.fcn_2d.FCN16s(*args: Any, **kwargs: Any)[source]

Bases: Module

forward(x)[source]
class elektronn3.models.fcn_2d.FCN32s(*args: Any, **kwargs: Any)[source]

Bases: Module

forward(x)[source]
class elektronn3.models.fcn_2d.FCN8s(*args: Any, **kwargs: Any)[source]

Bases: Module

forward(x)[source]
class elektronn3.models.fcn_2d.FCNs(*args: Any, **kwargs: Any)[source]

Bases: Module

forward(x)[source]
class elektronn3.models.fcn_2d.VGGNet(*args: Any, **kwargs: Any)[source]

Bases: VGG

forward(x)[source]
elektronn3.models.fcn_2d.make_layers(cfg, batch_norm=False, in_channels=3)[source]
elektronn3.models.fcn_2d.resize_conv_ala_distill(in_feat, out_feat, kernel_size, stride, padding, output_padding, dilation)[source]

# TODO: needs refinement to work with arbitrary kernel size, stride and padding etc. https://distill.pub/2016/deconv-checkerboard/ https://github.com/junyanz/pytorch-CycleGAN-and-pix2pix/issues/190 :param in_feat (): :param out_feat (): :param kernel_size (): :param stride (): :param padding (): :param output_padding (): :param dilation ():

Returns: