Utils uploaded

utils/arch_utils.py

@@ -0,0 +1,309 @@
+import math
+import torch
+
+from torch import nn as nn
+from torch.nn import functional as F
+from torch.nn import init as init
+from torch.nn.modules.batchnorm import _BatchNorm
+
+
+@torch.no_grad()
+def default_init_weights(module_list, scale=1, bias_fill=0, **kwargs):
+    """Initialize network weights.
+
+    Args:
+        module_list (list[nn.Module] | nn.Module): Modules to be initialized.
+        scale (float): Scale initialized weights, especially for residual
+            blocks. Default: 1.
+        bias_fill (float): The value to fill bias. Default: 0
+        kwargs (dict): Other arguments for initialization function.
+    """
+    if not isinstance(module_list, list):
+        module_list = [module_list]
+    for module in module_list:
+        for m in module.modules():
+            if isinstance(m, nn.Conv2d):
+                init.kaiming_normal_(m.weight, **kwargs)
+                m.weight.data *= scale
+                if m.bias is not None:
+                    m.bias.data.fill_(bias_fill)
+            elif isinstance(m, nn.Linear):
+                init.kaiming_normal_(m.weight, **kwargs)
+                m.weight.data *= scale
+                if m.bias is not None:
+                    m.bias.data.fill_(bias_fill)
+            elif isinstance(m, _BatchNorm):
+                init.constant_(m.weight, 1)
+                if m.bias is not None:
+                    m.bias.data.fill_(bias_fill)
+
+
+def make_layer(basic_block, num_basic_block, **kwarg):
+    """Make layers by stacking the same blocks.
+
+    Args:
+        basic_block (nn.module): nn.module class for basic block.
+        num_basic_block (int): number of blocks.
+
+    Returns:
+        nn.Sequential: Stacked blocks in nn.Sequential.
+    """
+    layers = []
+    for _ in range(num_basic_block):
+        layers.append(basic_block(**kwarg))
+    return nn.Sequential(*layers)
+
+
+class ResidualBlockNoBN(nn.Module):
+    """Residual block without BN.
+
+    It has a style of:
+        ---Conv-ReLU-Conv-+-
+         |________________|
+
+    Args:
+        num_feat (int): Channel number of intermediate features.
+            Default: 64.
+        res_scale (float): Residual scale. Default: 1.
+        pytorch_init (bool): If set to True, use pytorch default init,
+            otherwise, use default_init_weights. Default: False.
+    """
+
+    def __init__(self, num_feat=64, res_scale=1, pytorch_init=False):
+        super(ResidualBlockNoBN, self).__init__()
+        self.res_scale = res_scale
+        self.conv1 = nn.Conv2d(num_feat, num_feat, 3, 1, 1, bias=True)
+        self.conv2 = nn.Conv2d(num_feat, num_feat, 3, 1, 1, bias=True)
+        self.relu = nn.ReLU(inplace=True)
+
+        if not pytorch_init:
+            default_init_weights([self.conv1, self.conv2], 0.1)
+
+    def forward(self, x):
+        identity = x
+        out = self.conv2(self.relu(self.conv1(x)))
+        return identity + out * self.res_scale
+
+
+class Upsample(nn.Sequential):
+    """Upsample module.
+
+    Args:
+        scale (int): Scale factor. Supported scales: 2^n and 3.
+        num_feat (int): Channel number of intermediate features.
+    """
+
+    def __init__(self, scale, num_feat):
+        m = []
+        if (scale & (scale - 1)) == 0:  # scale = 2^n
+            for _ in range(int(math.log(scale, 2))):
+                m.append(nn.Conv2d(num_feat, 4 * num_feat, 3, 1, 1))
+                m.append(nn.PixelShuffle(2))
+        elif scale == 3:
+            m.append(nn.Conv2d(num_feat, 9 * num_feat, 3, 1, 1))
+            m.append(nn.PixelShuffle(3))
+        else:
+            raise ValueError(f'scale {scale} is not supported. '
+                             'Supported scales: 2^n and 3.')
+        super(Upsample, self).__init__(*m)
+
+
+def flow_warp(x,
+              flow,
+              interp_mode='bilinear',
+              padding_mode='zeros',
+              align_corners=True):
+    """Warp an image or feature map with optical flow.
+
+    Args:
+        x (Tensor): Tensor with size (n, c, h, w).
+        flow (Tensor): Tensor with size (n, h, w, 2), normal value.
+        interp_mode (str): 'nearest' or 'bilinear'. Default: 'bilinear'.
+        padding_mode (str): 'zeros' or 'border' or 'reflection'.
+            Default: 'zeros'.
+        align_corners (bool): Before pytorch 1.3, the default value is
+            align_corners=True. After pytorch 1.3, the default value is
+            align_corners=False. Here, we use the True as default.
+
+    Returns:
+        Tensor: Warped image or feature map.
+    """
+    assert x.size()[-2:] == flow.size()[1:3]
+    _, _, h, w = x.size()
+    # create mesh grid
+    grid_y, grid_x = torch.meshgrid(
+        torch.arange(0, h).type_as(x),
+        torch.arange(0, w).type_as(x))
+    grid = torch.stack((grid_x, grid_y), 2).float()  # W(x), H(y), 2
+    grid.requires_grad = False
+
+    vgrid = grid + flow
+    # scale grid to [-1,1]
+    vgrid_x = 2.0 * vgrid[:, :, :, 0] / max(w - 1, 1) - 1.0
+    vgrid_y = 2.0 * vgrid[:, :, :, 1] / max(h - 1, 1) - 1.0
+    vgrid_scaled = torch.stack((vgrid_x, vgrid_y), dim=3)
+    output = F.grid_sample(
+        x,
+        vgrid_scaled,
+        mode=interp_mode,
+        padding_mode=padding_mode,
+        align_corners=align_corners)
+
+    # TODO, what if align_corners=False
+    return output
+
+
+def resize_flow(flow,
+                size_type,
+                sizes,
+                interp_mode='bilinear',
+                align_corners=False):
+    """Resize a flow according to ratio or shape.
+
+    Args:
+        flow (Tensor): Precomputed flow. shape [N, 2, H, W].
+        size_type (str): 'ratio' or 'shape'.
+        sizes (list[int | float]): the ratio for resizing or the final output
+            shape.
+            1) The order of ratio should be [ratio_h, ratio_w]. For
+            downsampling, the ratio should be smaller than 1.0 (i.e., ratio
+            < 1.0). For upsampling, the ratio should be larger than 1.0 (i.e.,
+            ratio > 1.0).
+            2) The order of output_size should be [out_h, out_w].
+        interp_mode (str): The mode of interpolation for resizing.
+            Default: 'bilinear'.
+        align_corners (bool): Whether align corners. Default: False.
+
+    Returns:
+        Tensor: Resized flow.
+    """
+    _, _, flow_h, flow_w = flow.size()
+    if size_type == 'ratio':
+        output_h, output_w = int(flow_h * sizes[0]), int(flow_w * sizes[1])
+    elif size_type == 'shape':
+        output_h, output_w = sizes[0], sizes[1]
+    else:
+        raise ValueError(
+            f'Size type should be ratio or shape, but got type {size_type}.')
+
+    input_flow = flow.clone()
+    ratio_h = output_h / flow_h
+    ratio_w = output_w / flow_w
+    input_flow[:, 0, :, :] *= ratio_w
+    input_flow[:, 1, :, :] *= ratio_h
+    resized_flow = F.interpolate(
+        input=input_flow,
+        size=(output_h, output_w),
+        mode=interp_mode,
+        align_corners=align_corners)
+    return resized_flow
+
+
+# TODO: may write a cpp file
+def pixel_unshuffle(x, scale):
+    """ Pixel unshuffle.
+
+    Args:
+        x (Tensor): Input feature with shape (b, c, hh, hw).
+        scale (int): Downsample ratio.
+
+    Returns:
+        Tensor: the pixel unshuffled feature.
+    """
+    b, c, hh, hw = x.size()
+    out_channel = c * (scale**2)
+    assert hh % scale == 0 and hw % scale == 0
+    h = hh // scale
+    w = hw // scale
+    x_view = x.view(b, c, h, scale, w, scale)
+    return x_view.permute(0, 1, 3, 5, 2, 4).reshape(b, out_channel, h, w)
+
+
+
+class LayerNormFunction(torch.autograd.Function):
+
+    @staticmethod
+    def forward(ctx, x, weight, bias, eps):
+        ctx.eps = eps
+        N, C, H, W = x.size()
+        mu = x.mean(1, keepdim=True)
+        var = (x - mu).pow(2).mean(1, keepdim=True)
+        y = (x - mu) / (var + eps).sqrt()
+        ctx.save_for_backward(y, var, weight)
+        y = weight.view(1, C, 1, 1) * y + bias.view(1, C, 1, 1)
+        return y
+
+    @staticmethod
+    def backward(ctx, grad_output):
+        eps = ctx.eps
+
+        N, C, H, W = grad_output.size()
+        y, var, weight = ctx.saved_variables
+        g = grad_output * weight.view(1, C, 1, 1)
+        mean_g = g.mean(dim=1, keepdim=True)
+
+        mean_gy = (g * y).mean(dim=1, keepdim=True)
+        gx = 1. / torch.sqrt(var + eps) * (g - y * mean_gy - mean_g)
+        return gx, (grad_output * y).sum(dim=3).sum(dim=2).sum(dim=0), grad_output.sum(dim=3).sum(dim=2).sum(
+            dim=0), None
+
+class LayerNorm2d(nn.Module):
+
+    def __init__(self, channels, eps=1e-6):
+        super(LayerNorm2d, self).__init__()
+        self.register_parameter('weight', nn.Parameter(torch.ones(channels)))
+        self.register_parameter('bias', nn.Parameter(torch.zeros(channels)))
+        self.eps = eps
+
+    def forward(self, x):
+        return LayerNormFunction.apply(x, self.weight, self.bias, self.eps)
+
+# handle multiple input
+class MySequential(nn.Sequential):
+    def forward(self, *inputs):
+        for module in self._modules.values():
+            if type(inputs) == tuple:
+                inputs = module(*inputs)
+            else:
+                inputs = module(inputs)
+        return inputs
+
+import time
+def measure_inference_speed(model, data, max_iter=200, log_interval=50):
+    model.eval()
+
+    # the first several iterations may be very slow so skip them
+    num_warmup = 5
+    pure_inf_time = 0
+    fps = 0
+
+    # benchmark with 2000 image and take the average
+    for i in range(max_iter):
+
+        torch.cuda.synchronize()
+        start_time = time.perf_counter()
+
+        with torch.no_grad():
+            model(*data)
+
+        torch.cuda.synchronize()
+        elapsed = time.perf_counter() - start_time
+
+        if i >= num_warmup:
+            pure_inf_time += elapsed
+            if (i + 1) % log_interval == 0:
+                fps = (i + 1 - num_warmup) / pure_inf_time
+                print(
+                    f'Done image [{i + 1:<3}/ {max_iter}], '
+                    f'fps: {fps:.1f} img / s, '
+                    f'times per image: {1000 / fps:.1f} ms / img',
+                    flush=True)
+
+        if (i + 1) == max_iter:
+            fps = (i + 1 - num_warmup) / pure_inf_time
+            print(
+                f'Overall fps: {fps:.1f} img / s, '
+                f'times per image: {1000 / fps:.1f} ms / img',
+                flush=True)
+            break
+    return fps

utils/utils.py

@@ -0,0 +1,296 @@
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+import torch.nn.init as init
+
+from utils.arch_utils import LayerNorm2d
+
+def initialize_weights(net_l, scale=1):
+    if not isinstance(net_l, list):
+        net_l = [net_l]
+    for net in net_l:
+        for m in net.modules():
+            if isinstance(m, nn.Conv2d):
+                init.kaiming_normal_(m.weight, a=0, mode='fan_in')
+                m.weight.data *= scale  # for residual block
+                if m.bias is not None:
+                    m.bias.data.zero_()
+            elif isinstance(m, nn.Linear):
+                init.kaiming_normal_(m.weight, a=0, mode='fan_in')
+                m.weight.data *= scale
+                if m.bias is not None:
+                    m.bias.data.zero_()
+            elif isinstance(m, nn.BatchNorm2d):
+                init.constant_(m.weight, 1)
+                init.constant_(m.bias.data, 0.0)
+
+
+def make_layer(block, n_layers):
+    layers = []
+    for _ in range(n_layers):
+        layers.append(block())
+    return nn.Sequential(*layers)
+
+
+class ResidualBlock_noBN(nn.Module):
+    '''Residual block w/o BN
+    ---Conv-ReLU-Conv-+-
+     |________________|
+    '''
+
+    def __init__(self, nf=64):
+        super(ResidualBlock_noBN, self).__init__()
+        self.conv1 = nn.Conv2d(nf, nf, 3, 1, 1, bias=True)
+        self.conv2 = nn.Conv2d(nf, nf, 3, 1, 1, bias=True)
+
+        # initialization
+        initialize_weights([self.conv1, self.conv2], 0.1)
+
+    def forward(self, x):
+        identity = x
+        out = F.relu(self.conv1(x), inplace=True)
+        out = self.conv2(out)
+        return identity + out
+
+class ResidualBlock(nn.Module):
+    '''Residual block w/o BN
+    ---Conv-ReLU-Conv-+-
+     |________________|
+    '''
+
+    def __init__(self, nf=64):
+        super(ResidualBlock, self).__init__()
+        self.conv1 = nn.Conv2d(nf, nf, 3, 1, 1, bias=True)
+        self.bn = nn.BatchNorm2d(nf)
+        self.conv2 = nn.Conv2d(nf, nf, 3, 1, 1, bias=True)
+
+        # initialization
+        initialize_weights([self.conv1, self.conv2], 0.1)
+
+    def forward(self, x):
+        identity = x
+        out = F.relu(self.bn(self.conv1(x)), inplace=True)
+        out = self.conv2(out)
+        return identity + out
+
+###########################################################################################################
+
+
+class SimpleGate(nn.Module):
+    def forward(self, x):
+        x1, x2 = x.chunk(2, dim=1)
+        return x1 * x2
+
+class SGE(nn.Module):
+    def __init__(self, dw_channel):
+        super().__init__() 
+        self.dwc = nn.Conv2d(in_channels=dw_channel //2, out_channels=dw_channel//2, kernel_size=3, padding=1, stride=1, groups=dw_channel//2, bias=True)
+    def forward(self, x):
+        x1, x2 = x.chunk(2, dim=1)
+        x1 = self.dwc(x1)
+        return x1 * x2
+    
+class SpaBlock(nn.Module):
+    def __init__(self, nc, DW_Expand = 2,  FFN_Expand=2, drop_out_rate=0.):
+        super(SpaBlock, self).__init__()
+        dw_channel = nc * DW_Expand
+        self.conv1 = nn.Conv2d(in_channels=nc, out_channels=dw_channel, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        self.conv2 = nn.Conv2d(in_channels=dw_channel, out_channels=dw_channel, kernel_size=3, padding=1, stride=1, groups=dw_channel,
+                               bias=True) # the dconv
+        self.conv3 = nn.Conv2d(in_channels=dw_channel // 2, out_channels=nc, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        
+        # Simplified Channel Attention
+        self.sca = nn.Sequential(
+            nn.AdaptiveAvgPool2d(1),
+            nn.Conv2d(in_channels=dw_channel // 2, out_channels=dw_channel // 2, kernel_size=1, padding=0, stride=1,
+                      groups=1, bias=True),
+        )
+
+        # SimpleGate
+        self.sg = SimpleGate()
+
+        ffn_channel = FFN_Expand * nc
+        self.conv4 = nn.Conv2d(in_channels=nc, out_channels=ffn_channel, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        self.conv5 = nn.Conv2d(in_channels=ffn_channel // 2, out_channels=nc, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+
+        self.norm1 = LayerNorm2d(nc)
+        self.norm2 = LayerNorm2d(nc)
+
+        self.dropout1 = nn.Dropout(drop_out_rate) if drop_out_rate > 0. else nn.Identity()
+        self.dropout2 = nn.Dropout(drop_out_rate) if drop_out_rate > 0. else nn.Identity()
+
+        self.beta = nn.Parameter(torch.zeros((1, nc, 1, 1)), requires_grad=True)
+        self.gamma = nn.Parameter(torch.zeros((1, nc, 1, 1)), requires_grad=True)
+
+    def forward(self, x):
+
+        x = self.norm1(x) # size [B, C, H, W]
+
+        x = self.conv1(x) # size [B, 2*C, H, W]
+        x = self.conv2(x) # size [B, 2*C, H, W]
+        x = self.sg(x)    # size [B, C, H, W]
+        x = x * self.sca(x) # size [B, C, H, W]
+        x = self.conv3(x) # size [B, C, H, W]
+
+        x = self.dropout1(x)
+
+        y = x + x * self.beta # size [B, C, H, W]
+
+        x = self.conv4(self.norm2(y)) # size [B, 2*C, H, W]
+        x = self.sg(x)  # size [B, C, H, W]
+        x = self.conv5(x) # size [B, C, H, W]
+
+        x = self.dropout2(x)
+
+        return y + x * self.gamma
+
+class FreBlock(nn.Module):
+    def __init__(self, nc):
+        super(FreBlock, self).__init__()
+        self.fpre = nn.Conv2d(nc, nc, 1, 1, 0)
+        self.process1 = nn.Sequential(
+            nn.Conv2d(nc, nc, 1, 1, 0),
+            nn.LeakyReLU(0.1, inplace=True),
+            nn.Conv2d(nc, nc, 1, 1, 0))
+        self.process2 = nn.Sequential(
+            nn.Conv2d(nc, nc, 1, 1, 0),
+            nn.LeakyReLU(0.1, inplace=True),
+            nn.Conv2d(nc, nc, 1, 1, 0))
+
+    def forward(self, x):
+        _, _, H, W = x.shape
+        x_freq = torch.fft.rfft2(self.fpre(x), norm='backward')
+        mag = torch.abs(x_freq)
+        pha = torch.angle(x_freq)
+        mag = self.process1(mag)
+        pha = self.process2(pha)
+        real = mag * torch.cos(pha)
+        imag = mag * torch.sin(pha)
+        x_out = torch.complex(real, imag)
+        x_out = torch.fft.irfft2(x_out, s=(H, W), norm='backward')
+
+        return x_out+x
+
+    
+class SFBlock(nn.Module):
+    def __init__(self, nc, DW_Expand = 2,  FFN_Expand=2):
+        super(SFBlock, self).__init__()
+        dw_channel = nc * DW_Expand
+        self.conv1 = nn.Conv2d(in_channels=nc, out_channels=dw_channel, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        self.conv2 = nn.Conv2d(in_channels=dw_channel, out_channels=dw_channel, kernel_size=3, padding=1, stride=1, groups=dw_channel,
+                               bias=True) # the dconv
+        self.conv3 = nn.Conv2d(in_channels=dw_channel // 2, out_channels=nc, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        
+        self.fatt = FreBlock(dw_channel // 2)
+        self.sge = SGE(dw_channel)
+
+        # SimpleGate
+        self.sg = SimpleGate()
+
+        ffn_channel = FFN_Expand * nc
+        self.conv4 = nn.Conv2d(in_channels=nc, out_channels=ffn_channel, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+        self.conv5 = nn.Conv2d(in_channels=ffn_channel // 2, out_channels=nc, kernel_size=1, padding=0, stride=1, groups=1, bias=True)
+
+        self.norm1 = LayerNorm2d(nc)
+        self.norm2 = LayerNorm2d(nc)
+
+        self.beta = nn.Parameter(torch.zeros((1, nc, 1, 1)), requires_grad=True)
+        self.gamma = nn.Parameter(torch.zeros((1, nc, 1, 1)), requires_grad=True)
+
+    def forward(self, x):
+
+        x = self.norm1(x) # size [B, C, H, W]
+
+        x = self.conv1(x) # size [B, 2*C, H, W]
+        x = self.conv2(x) # size [B, 2*C, H, W]
+        x = self.sge(x)    # size [B, C, H, W]
+      
+        x = self.fatt(x)
+        x = self.conv3(x) # size [B, C, H, W]
+
+        y = x + x * self.beta # size [B, C, H, W]
+
+        x = self.conv4(self.norm2(y)) # size [B, 2*C, H, W]
+        x = self.sg(x)  # size [B, C, H, W]
+        x = self.conv5(x) # size [B, C, H, W]
+
+        return y + x * self.gamma
+    
+class ProcessBlock(nn.Module):
+    def __init__(self, in_nc, spatial = True):
+        super(ProcessBlock,self).__init__()
+        self.spatial = spatial
+        self.spatial_process = SpaBlock(in_nc) if spatial else nn.Identity()
+        self.frequency_process = FreBlock(in_nc)
+        self.cat = nn.Conv2d(2*in_nc,in_nc,1,1,0) if spatial else nn.Conv2d(in_nc,in_nc,1,1,0)
+
+    def forward(self, x):
+        xori = x
+        x_freq = self.frequency_process(x)
+        x_spatial = self.spatial_process(x)
+        xcat = torch.cat([x_spatial,x_freq],1)
+        x_out = self.cat(xcat) if self.spatial else self.cat(x_freq)
+
+        return x_out+xori
+
+class SFNet(nn.Module):
+
+    def __init__(self, nc,n=5):
+        super(SFNet,self).__init__()
+
+        self.list_block = list()
+        for index in range(n):
+
+            self.list_block.append(ProcessBlock(nc,spatial=False))
+  
+        self.block = nn.Sequential(*self.list_block)
+
+    def forward(self, x):
+
+        x_ori = x
+        x_out = self.block(x_ori)
+        xout = x_ori + x_out
+
+        return xout
+
+class AmplitudeNet_skip(nn.Module):
+    def __init__(self, nc,n=1):
+        super(AmplitudeNet_skip,self).__init__()
+        
+        self.conv_init = nn.Conv2d(3, nc, 1, 1, 0)
+        self.conv1 = SFBlock (nc)
+        self.conv2 = SFBlock (nc)
+        self.conv3 = SFBlock (nc)
+        self.conv_out = nn.Conv2d(nc, 3, 1, 1, 0)
+
+    def forward(self, x):
+        
+        x_lr = F.interpolate(x, scale_factor=0.5, mode='bilinear') # Resize and Normalize SNR map
+        
+        x_lr = self.conv_init(x_lr)
+        x_lr = self.conv1(x_lr)
+        x_lr = self.conv2(x_lr)
+        x_lr = self.conv3(x_lr)
+        x_lr = self.conv_out(x_lr)
+        
+        xout = F.interpolate(x_lr, scale_factor=2, mode='bilinear') # Resize and Normalize SNR map
+        
+        return xout
+
+    
+###########################################################################################################
+
+class SG(nn.Module):
+    def forward(self, x):
+        x1, x2 = x.chunk(2, dim=1)
+        return x1 * x2
+    
+
+class SGE(nn.Module):
+    def __init__(self, dw_channel):
+        super().__init__() 
+        self.dwc = nn.Conv2d(in_channels=dw_channel //2, out_channels=dw_channel//2, kernel_size=3, padding=1, stride=1, groups=dw_channel//2, bias=True)
+    def forward(self, x):
+        x1, x2 = x.chunk(2, dim=1)
+        x1 = self.dwc(x1)
+        return x1 * x2