def __init__(self, ch):
super(SNResFCBlock, self).__init__()
initializer = chainer.initializers.GlorotUniform()
with self.init_scope():
self.l0 = SNLinear(ch, ch, initialW=initializer)
self.l1 = SNLinear(ch, ch, initialW=initializer)
self.bn0 = L.BatchNormalization(ch)
self.bn1 = L.BatchNormalization(ch)
def __init__(self, ch=64, activation=F.relu):
super(SNResNetDiscriminator, self).__init__()
self.activation = activation
initializer = chainer.initializers.GlorotUniform()
#w = chainer.initializers.Normal(0.02)
with self.init_scope():
self.block1 = OptimizedBlock(3, ch)
#self.block1 = L.Convolution2D(3, ch // 8, 1, 1, 0, initialW=w)
self.block2 = BlockDisc(ch,
ch * 2,
activation=activation,
downsample=True)
self.block3 = BlockDisc(ch * 2,
ch * 4,
activation=activation,
downsample=True)
self.block4 = BlockDisc(ch * 4,
ch * 8,
activation=activation,
downsample=True)
self.block5 = BlockDisc(ch * 8,
ch * 8,
activation=activation,
downsample=False)
self.l6 = SNLinear(ch * 8, 1, initialW=initializer)
def __init__(self, output_dim=1, wscale=0.02, ch=512, bottom_width=4):
w = chainer.initializers.Normal(wscale)
super(SNMnistDiscriminator, self).__init__()
with self.init_scope():
self.c0 = SNConvolution2D(1,
ch // 4,
ksize=4,
stride=2,
pad=1,
initialW=w)
self.c1 = SNConvolution2D(ch // 4,
ch // 2,
ksize=4,
stride=2,
pad=2,
initialW=w)
self.c2 = SNConvolution2D(ch // 2,
ch,
ksize=4,
stride=2,
pad=1,
initialW=w)
self.lout = SNLinear(bottom_width * bottom_width * ch,
output_dim,
initialW=w)
def __init__(self, ch=512, wscale=0.02):
w = chainer.initializers.Normal(wscale)
super(SNPGGANDiscriminator, self).__init__()
with self.init_scope():
self.in1 = SNConvolution2D(3, ch // 8, 1, 1, 0, initialW=w)
self.b1 = SNDiscriminatorBlock(ch // 8, ch // 4, initialW=w)
self.b2 = SNDiscriminatorBlock(ch // 4, ch // 2, initialW=w)
self.b3 = SNDiscriminatorBlock(ch // 2, ch, initialW=w)
self.out0 = SNConvolution2D(ch, ch, 3, 1, 1, initialW=w)
self.out1 = SNConvolution2D(ch, ch, 4, 1, 0, initialW=w)
self.out2 = SNLinear(ch, 1, initialW=w)
def __init__(self, ch=64, activation=F.relu):
super(Discriminator48, self).__init__()
self.activation = activation
initializer = chainer.initializers.GlorotUniform()
with self.init_scope():
self.block1 = OptimizedBlock(3, ch * 1)
self.block2 = DiscriminatorBlock(ch * 1, ch * 2, activation=activation, downsample=True)
self.block3 = DiscriminatorBlock(ch * 2, ch * 4, activation=activation, downsample=True)
self.block4 = DiscriminatorBlock(ch * 4, ch * 8, activation=activation, downsample=True)
self.block5 = DiscriminatorBlock(ch * 8, ch * 16, activation=activation, downsample=False)
self.l5 = SNLinear(ch * 16, 1, initialW=initializer)
def __init__(self, bottom_width=8, ch=128, ouput_dim=1):
w = chainer.initializers.HeNormal()
super(SNResnetDiscriminator, self).__init__()
self.bottom_width = bottom_width
self.ch = ch
with self.init_scope():
self.r0 = SNOptimizedResBlock1(128)
self.r1 = SNDownResBlock2(128)
self.r2 = SNResBlock3(128)
self.r3 = SNResBlock3(128)
self.l4 = SNLinear(bottom_width * bottom_width * ch, ouput_dim, initialW=w)
def __init__(self, ch=64, activation=F.relu,sn=True):
super(Discriminator32, self).__init__()
self.activation = activation
initializer = chainer.initializers.GlorotUniform()
with self.init_scope():
self.block1 = OptimizedBlock(3, ch*2,sn=sn)
self.block2 = DiscriminatorBlock(ch * 2, ch * 2, activation=activation, downsample=True,sn=sn)
self.block3 = DiscriminatorBlock(ch * 2, ch * 2, activation=activation, downsample=False,sn=sn)
self.block4 = DiscriminatorBlock(ch * 2, ch * 2, activation=activation, downsample=False,sn=sn)
if sn:
self.l5 = SNLinear(ch * 2, 1, initialW=initializer)
else:
self.l5 = L.Linear(ch * 2, 1, initialW=initializer)
def __init__(self, bottom_width=4, ch=512, wscale=0.02, output_dim=1):
w = chainer.initializers.Normal(wscale)
super(SNDCGANDiscriminator, self).__init__()
with self.init_scope():
self.c0_0 = SNConvolution2D(3, ch // 8, 3, 1, 1, initialW=w)
self.c0_1 = SNConvolution2D(ch // 8, ch // 4, 4, 2, 1, initialW=w)
self.c1_0 = SNConvolution2D(ch // 4, ch // 4, 3, 1, 1, initialW=w)
self.c1_1 = SNConvolution2D(ch // 4, ch // 2, 4, 2, 1, initialW=w)
self.c2_0 = SNConvolution2D(ch // 2, ch // 2, 3, 1, 1, initialW=w)
self.c2_1 = SNConvolution2D(ch // 2, ch // 1, 4, 2, 1, initialW=w)
self.c3_0 = SNConvolution2D(ch // 1, ch // 1, 3, 1, 1, initialW=w)
self.l4 = SNLinear(bottom_width * bottom_width * ch,
output_dim,
initialW=w)
def __init__(self, ch=512, wscale=0.02, sn=True):
super(Discriminator32, self).__init__()
w = chainer.initializers.Normal(wscale)
with self.init_scope():
if sn:
print('sn')
self.in_ = SNConvolution2D(3, ch // 8, 1, 1, 0, initialW=w)
else:
print('sn_free')
self.in_ = L.Convolution2D(3, ch // 8, 1, 1, 0, initialW=w)
self.b3 = DiscriminatorBlock(ch // 8, ch // 4, initialW=w, sn=sn)
self.b2 = DiscriminatorBlock(ch // 4, ch // 2, initialW=w, sn=sn)
self.b1 = DiscriminatorBlock(ch // 2, ch, initialW=w, sn=sn)
if sn:
self.out0 = SNConvolution2D(ch, ch, 3, 1, 1, initialW=w)
self.out1 = SNConvolution2D(ch, ch, 4, 1, 0, initialW=w)
self.out2 = SNLinear(ch, 1, initialW=w)
else:
self.out0 = L.Convolution2D(ch, ch, 3, 1, 1, initialW=w)
self.out1 = L.Convolution2D(ch, ch, 4, 1, 0, initialW=w)
self.out2 = L.Linear(ch, 1, initialW=w)