Finish up spsr_switched

This commit is contained in:
James Betker 2020-08-07 21:03:48 -06:00
parent 1d5f4f6102
commit 887806ffa0
2 changed files with 9 additions and 4 deletions

View File

@ -7,6 +7,7 @@ from .RRDBNet_arch import RRDB
from models.archs.arch_util import ConvGnLelu, UpconvBlock
from models.archs.SwitchedResidualGenerator_arch import MultiConvBlock, ConvBasisMultiplexer, ConfigurableSwitchComputer
from switched_conv_util import save_attention_to_image_rgb
from switched_conv import compute_attention_specificity
import functools
@ -428,7 +429,7 @@ class SPSRNetSimplifiedNoSkip(nn.Module):
class SwitchedSpsr(nn.Module):
def __init__(self, in_nc, out_nc, nf, nb, upscale=4):
def __init__(self, in_nc, out_nc, nf, upscale=4):
super(SwitchedSpsr, self).__init__()
n_upscale = int(math.log(upscale, 2))
@ -442,7 +443,7 @@ class SwitchedSpsr(nn.Module):
switch_processing_layers, trans_counts)
pretransform_fn = functools.partial(ConvGnLelu, transformation_filters, transformation_filters, norm=False, bias=False, weight_init_factor=.1)
transform_fn = functools.partial(MultiConvBlock, transformation_filters, int(transformation_filters * 1.5),
transformation_filters, kernel_size=3, depth=trans_layers,
transformation_filters, kernel_size=3, depth=3,
weight_init_factor=.1)
# Feature branch
@ -512,11 +513,13 @@ class SwitchedSpsr(nn.Module):
x_out_branch = self.grad_branch_output_conv(x_grad)
x__branch_pretrain_cat = torch.cat([x_grad, x_fea], dim=1)
x__branch_pretrain_cat = self._branch_pretrain_concat(x__branch_pretrain_cat)
x__branch_pretrain_cat, a4 = self._branch_pretrain_sw(x__branch_pretrain_cat, True)
x_out = self._branch_pretrain_concat(x__branch_pretrain_cat)
x_out = self._branch_pretrain_HR_conv0(x_out)
x_out = self._branch_pretrain_HR_conv0(x__branch_pretrain_cat)
x_out = self._branch_pretrain_HR_conv1(x_out)
self.attentions = [a1, a2, a3, a4]
return x_out_branch, x_out, x_grad
def set_temperature(self, temp):

View File

@ -111,6 +111,8 @@ def define_G(opt, net_key='network_G'):
elif which_model == 'spsr_net_improved_noskip':
netG = spsr.SPSRNetSimplifiedNoSkip(in_nc=opt_net['in_nc'], out_nc=opt_net['out_nc'], nf=opt_net['nf'],
nb=opt_net['nb'], upscale=opt_net['scale'])
elif which_model == "spsr_switched":
netG = spsr.SwitchedSpsr(in_nc=3, out_nc=3, nf=opt_net['nf'], upscale=opt_net['scale'])
# image corruption
elif which_model == 'HighToLowResNet':