From 645d0ca767ca280a1711ee1924a1c8ce994f1632 Mon Sep 17 00:00:00 2001 From: James Betker Date: Wed, 17 Jun 2020 17:18:28 -0600 Subject: [PATCH] ResidualGen mods - Add filters_mid spec which allows a expansion->squeeze for the transformation layers. - Add scale and bias AFTER the switch - Remove identity transform (models were converging on this) - Move attention image generation and temperature setting into new function which gets called every step with a save path --- codes/models/SRGAN_model.py | 4 ++ .../archs/SwitchedResidualGenerator_arch.py | 47 +++++++++++-------- codes/models/networks.py | 1 + 3 files changed, 32 insertions(+), 20 deletions(-) diff --git a/codes/models/SRGAN_model.py b/codes/models/SRGAN_model.py index e3ce73eb..9a9daff8 100644 --- a/codes/models/SRGAN_model.py +++ b/codes/models/SRGAN_model.py @@ -183,6 +183,10 @@ class SRGANModel(BaseModel): self.pix = [t.to(self.device) for t in torch.chunk(data['PIX'], chunks=self.mega_batch_factor, dim=0)] def optimize_parameters(self, step): + # Some generators have variants depending on the current step. + if hasattr(self.netG.module, "update_for_step"): + self.netG.module.update_for_step(step, os.path.join(self.opt['path']['models'], "..")) + # G for p in self.netD.parameters(): p.requires_grad = False diff --git a/codes/models/archs/SwitchedResidualGenerator_arch.py b/codes/models/archs/SwitchedResidualGenerator_arch.py index a2a41c99..4f55948d 100644 --- a/codes/models/archs/SwitchedResidualGenerator_arch.py +++ b/codes/models/archs/SwitchedResidualGenerator_arch.py @@ -29,11 +29,12 @@ class ConvBnLelu(nn.Module): class ResidualBranch(nn.Module): - def __init__(self, filters_in, filters_out, kernel_size, depth): + def __init__(self, filters_in, filters_mid, filters_out, kernel_size, depth): + assert depth >= 2 super(ResidualBranch, self).__init__() - self.bnconvs = nn.ModuleList([ConvBnLelu(filters_in, filters_out, kernel_size)] + - [ConvBnLelu(filters_out, filters_out, kernel_size) for i in range(depth-2)] + - [ConvBnLelu(filters_out, filters_out, kernel_size, lelu=False)]) + self.bnconvs = nn.ModuleList([ConvBnLelu(filters_in, filters_mid, kernel_size)] + + [ConvBnLelu(filters_mid, filters_mid, kernel_size) for i in range(depth-2)] + + [ConvBnLelu(filters_mid, filters_out, kernel_size, lelu=False)]) self.scale = nn.Parameter(torch.ones(1)) self.bias = nn.Parameter(torch.zeros(1)) @@ -67,16 +68,15 @@ class SwitchComputer(nn.Module): self.proc_switch_conv = ConvBnLelu(final_filters, proc_block_filters) self.final_switch_conv = nn.Conv2d(proc_block_filters, transform_count, 1, 1, 0) - # Always include the identity transform (all zeros), hence transform_count-10 - self.transforms = nn.ModuleList([transform_block() for i in range(transform_count-1)]) + self.transforms = nn.ModuleList([transform_block() for i in range(transform_count)]) - # And the switch itself + # And the switch itself, including learned scalars self.switch = BareConvSwitch(initial_temperature=init_temp) + self.scale = nn.Parameter(torch.ones(1)) + self.bias = nn.Parameter(torch.zeros(1)) def forward(self, x, output_attention_weights=False): xformed = [t.forward(x) for t in self.transforms] - # Append the identity transform. - xformed.append(torch.zeros_like(xformed[0])) multiplexer = self.filter_conv(x) for block in self.reduction_blocks: @@ -88,18 +88,23 @@ class SwitchComputer(nn.Module): # Interpolate the multiplexer across the entire shape of the image. multiplexer = F.interpolate(multiplexer, size=x.shape[2:], mode='nearest') - return self.switch(xformed, multiplexer, output_attention_weights) + outputs, attention = self.switch(xformed, multiplexer, True) + outputs = outputs * self.scale + self.bias + if output_attention_weights: + return outputs, attention + else: + return outputs def set_temperature(self, temp): self.switch.set_attention_temperature(temp) class ConfigurableSwitchedResidualGenerator(nn.Module): - def __init__(self, switch_filters, switch_reductions, switch_processing_layers, trans_counts, trans_kernel_sizes, trans_layers, initial_temp=20, final_temperature_step=50000): + def __init__(self, switch_filters, switch_reductions, switch_processing_layers, trans_counts, trans_kernel_sizes, trans_layers, trans_filters_mid, initial_temp=20, final_temperature_step=50000): super(ConfigurableSwitchedResidualGenerator, self).__init__() switches = [] - for filters, sw_reduce, sw_proc, trans_count, kernel, layers in zip(switch_filters, switch_reductions, switch_processing_layers, trans_counts, trans_kernel_sizes, trans_layers): - switches.append(SwitchComputer(3, filters, functools.partial(ResidualBranch, 3, 3, kernel_size=kernel, depth=layers), trans_count, sw_reduce, sw_proc, initial_temp)) + for filters, sw_reduce, sw_proc, trans_count, kernel, layers, mid_filters in zip(switch_filters, switch_reductions, switch_processing_layers, trans_counts, trans_kernel_sizes, trans_layers, trans_filters_mid): + switches.append(SwitchComputer(3, filters, functools.partial(ResidualBranch, 3, mid_filters, 3, kernel_size=kernel, depth=layers), trans_count, sw_reduce, sw_proc, initial_temp)) initialize_weights(switches, 1) # Initialize the transforms with a lesser weight, since they are repeatedly added on to the resultant image. initialize_weights([s.transforms for s in switches], .2 / len(switches)) @@ -107,6 +112,7 @@ class ConfigurableSwitchedResidualGenerator(nn.Module): self.transformation_counts = trans_counts self.init_temperature = initial_temp self.final_temperature_step = final_temperature_step + self.attentions = None def forward(self, x): self.attentions = [] @@ -119,14 +125,15 @@ class ConfigurableSwitchedResidualGenerator(nn.Module): def set_temperature(self, temp): [sw.set_temperature(temp) for sw in self.switches] + def update_for_step(self, step, experiments_path='.'): + if self.attentions: + temp = max(1, int(self.init_temperature * (self.final_temperature_step - step) / self.final_temperature_step)) + self.set_temperature(temp) + if step % 2 == 0: + [save_attention_to_image(experiments_path, self.attentions[i], self.transformation_counts[i], step, "a%i" % (i+1,), l_mult=float(self.transformation_counts[i]/4)) for i in range(len(self.switches))] + def get_debug_values(self, step): - # Take the chance to update the temperature here. - temp = max(1, int(self.init_temperature * (self.final_temperature_step - step) / self.final_temperature_step)) - self.set_temperature(temp) - - if step % 250 == 0: - [save_attention_to_image(self.attentions[i], self.transformation_counts[i], step, "a%i" % (i+1,), l_mult=float(self.transformation_counts[i]/4)) for i in range(len(self.switches))] - + temp = self.switches[0].switch.temperature mean_hists = [compute_attention_specificity(att, 2) for att in self.attentions] means = [i[0] for i in mean_hists] hists = [i[1].clone().detach().cpu().flatten() for i in mean_hists] diff --git a/codes/models/networks.py b/codes/models/networks.py index 255c10cc..e5cccbcf 100644 --- a/codes/models/networks.py +++ b/codes/models/networks.py @@ -66,6 +66,7 @@ def define_G(opt, net_key='network_G'): netG = SwitchedGen_arch.ConfigurableSwitchedResidualGenerator(switch_filters=opt_net['switch_filters'], switch_reductions=opt_net['switch_reductions'], switch_processing_layers=opt_net['switch_processing_layers'], trans_counts=opt_net['trans_counts'], trans_kernel_sizes=opt_net['trans_kernel_sizes'], trans_layers=opt_net['trans_layers'], + trans_filters_mid=opt_net['trans_filters_mid'], initial_temp=opt_net['temperature'], final_temperature_step=opt_net['temperature_final_step']) # image corruption