def __init__(self, in_chans, out_chans, drop_prob):
"""
Args:
in_chans (int): Number of channels in the input.
out_chans (int): Number of channels in the output.
drop_prob (float): Dropout probability.
"""
super().__init__()
self.in_chans = int(in_chans)
self.out_chans = int(out_chans)
self.drop_prob = drop_prob
self.layers = ComplexSequential(
ComplexConv2d(in_chans,
out_chans,
kernel_size=3,
padding=1,
bias=False,
setW=False), ComplexBatchNorm2d(out_chans),
ComplexReLU(), ComplexDropout2d(p=drop_prob, inplace=True),
ComplexConv2d(out_chans,
out_chans,
kernel_size=3,
padding=1,
bias=False,
setW=False), ComplexBatchNorm2d(out_chans),
ComplexReLU(), ComplexDropout2d(p=drop_prob, inplace=True))
def __init__(self,
in_chans=1,
out_chans=1,
drop_prob=0.05,
resolution=128):
"""
Args:
in_chans (int): Number of channels in the input.
out_chans (int): Number of channels in the output.
"""
super().__init__()
self.in_chans = int(in_chans)
self.out_chans = int(out_chans)
self.drop_prob = drop_prob
self.resolution = resolution
self.c_layer1 = ComplexConv2d(in_channels=1,
out_channels=resolution,
kernel_size=(1, resolution),
padding=(0, 0),
stride=1,
bias=False)
self.c_layer2 = ComplexConv2d(in_channels=1,
out_channels=resolution,
kernel_size=(1, resolution),
padding=(0, 0),
stride=1,
bias=False)
def __init__(self, in_chans, out_chans, drop_prob, chans, num_pool_layers, resolution):
super().__init__()
self.in_chans = in_chans
self.out_chans = out_chans
self.drop_prob = drop_prob
self.resolution= resolution
self.layer1=ComplexConv2d(in_channels=1, out_channels=resolution, kernel_size=(1,resolution),padding=(0,0), stride=1, bias=False)
self.layer2=ComplexConv2d(in_channels=1, out_channels=resolution, kernel_size=(1,resolution),padding=(0,0), stride=1, bias=False)
self.chans = chans
self.num_pool_layers = num_pool_layers
self.down_sample_layers = nn.ModuleList([ConvBlock(in_chans, chans, drop_prob)])
ch = chans
for i in range(num_pool_layers - 1):
self.down_sample_layers += [ConvBlock(ch, ch * 2, drop_prob)]
ch *= 2
self.conv = ConvBlock(ch, ch * 2, drop_prob)
self.up_conv = nn.ModuleList()
self.up_transpose_conv = nn.ModuleList()
for i in range(num_pool_layers - 1):
self.up_transpose_conv += [TransposeConvBlock(ch * 2, ch)]
self.up_conv += [ConvBlock(ch * 2, ch, drop_prob)]
ch //= 2
self.up_transpose_conv += [TransposeConvBlock(ch * 2, ch)]
self.up_conv += [
nn.Sequential(
ConvBlock(ch * 2, ch, drop_prob),
nn.Conv2d(ch, self.out_chans, kernel_size=1, stride=1),
)]
def __init__(self, k1=2, c1=40, k2=2, c2=100, d1=96, d2=10):
super(Encoder_f, self).__init__()
self.conv1 = ComplexConv2d(1, c1, k1, 1, padding=0)
self.bn = ComplexBatchNorm2d(c1)
self.conv2 = ComplexConv2d(c1, c2, k2, 1, padding=0)
self.c2 = c2
self.fc1 = ComplexLinear(2 * 2 * c2, d1)
self.fc2 = ComplexLinear(d1, d2)
def __init__(self, in_chans, out_chans, drop_prob, resolution):
super().__init__()
self.in_chans = in_chans
self.out_chans = out_chans
self.drop_prob = drop_prob
self.resolution= resolution
self.layer1=ComplexConv2d(in_channels=1, out_channels=resolution, kernel_size=(1,resolution),padding=(0,0), stride=1, bias=False)
self.layer2=ComplexConv2d(in_channels=1, out_channels=resolution, kernel_size=(1,resolution),padding=(0,0), stride=1, bias=False)
def __init__(self):
super(ComplexNet, self).__init__()
self.pool = ComplexMaxPool2d(kernel_size=2, stride=2)
self.conv1 = ComplexConv2d(1, 20, 5, 1)
self.bn = NaiveComplexBatchNorm2d(20)
self.conv2 = ComplexConv2d(20, 50, 5, 1)
self.fc1 = ComplexLinear(4 * 4 * 50, 500)
self.fc2 = ComplexLinear(500, 10)
def __init__(self, k1=2, c1=40, k2=2, c2=100, k3=3, c3=3, d1=96, d2=10):
super(Encoder_r, self).__init__()
self.conv1 = ComplexConv2d(1, c1, k1, 1, padding=0)
self.bn1 = ComplexBatchNorm2d(c1)
self.conv2 = ComplexConv2d(c1, c2, k2, 1, padding=0)
self.bn2 = ComplexBatchNorm2d(c2)
self.conv3 = ComplexConv2d(c2, c3, k3, 1, padding=0)
self.fc1 = ComplexLinear(2 * 2 * c3, d1)
self.fc2 = ComplexLinear(d1, d2)
self.c3 = c3
def __init__(self):
super(ComplexDiscriminator, self).__init__()
# torch.nn.Conv2d(in_channels, out_channels, kernel_size, stride=1,
# padding=0, dilation=1, groups=1, bias=True)
# input(N, C_in, H_in, W_in), output(N, C_out, H_out, W_out)
# H_out=[H_in + 2×padding[0] - dilation[0]×(kernel_size[0]−1) − 1]/stride[0] + 1
self.conv1 = ComplexConv2d(1, 64, 4, 2, 1)
self.conv2 = ComplexConv2d(64, 128, 4, 2, 1)
self.conv3 = ComplexConv2d(128, 256, 4, 2, 1)
self.conv4 = ComplexConv2d(256, 512, 4, 2, 1)
self.conv5 = ComplexConv2d(512, 1024, 4, 2, 1)
self.dense = ComplexLinear(1024 * 4 * 4, 1)
def __init__(self, in_chans, out_chans, chans, num_pool_layers, drop_prob):
"""
Args:
in_chans (int): Number of channels in the input to the U-Net model.
out_chans (int): Number of channels in the output to the U-Net model.
chans (int): Number of output channels of the first convolution layer.
num_pool_layers (int): Number of down-sampling and up-sampling layers.
drop_prob (float): Dropout probability.
"""
super().__init__()
self.in_chans = int(in_chans)
self.out_chans = int(out_chans)
self.chans = chans
self.num_pool_layers = num_pool_layers
self.drop_prob = drop_prob
self.down_sample_layers = nn.ModuleList([ComplexConvBlock(in_chans, chans, drop_prob)])
ch = chans
for i in range(num_pool_layers - 1):
self.down_sample_layers += [ComplexConvBlock(ch, ch * 2, drop_prob)]
ch *= 2
self.conv = ComplexConvBlock(ch, ch * 2, drop_prob)
self.up_conv = nn.ModuleList()
self.up_transpose_conv = nn.ModuleList()
for i in range(num_pool_layers - 1):
self.up_transpose_conv += [ComplexTransposeConvBlock(ch * 2, ch)]
self.up_conv += [ComplexConvBlock(ch * 2, ch, drop_prob)]
ch //= 2
self.up_transpose_conv += [ComplexTransposeConvBlock(ch * 2, ch)]
self.up_conv += [
ComplexSequential(
ComplexConvBlock(ch * 2, ch, drop_prob),
ComplexConv2d(ch, self.out_chans, kernel_size=1, stride=1),
)]