def __init__(self,
in_channels,
out_channels,
mid_channels,
use_bias,
use_bn,
data_format="channels_last",
**kwargs):
super(StemBlock, self).__init__(**kwargs)
self.conv1 = conv7x7_block(in_channels=in_channels,
out_channels=mid_channels,
strides=2,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="conv1")
self.pool1 = MaxPool2d(pool_size=3,
strides=2,
padding=0,
ceil_mode=True,
data_format=data_format,
name="pool1")
self.conv2 = Inception3x3Branch(in_channels=mid_channels,
out_channels=out_channels,
mid_channels=mid_channels,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="conv2")
self.pool2 = MaxPool2d(pool_size=3,
strides=2,
padding=0,
ceil_mode=True,
data_format=data_format,
name="pool2")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(InceptInitBlock, self).__init__(**kwargs)
assert (out_channels == 192)
self.conv1 = InceptConv(in_channels=in_channels,
out_channels=32,
kernel_size=3,
strides=2,
padding=0,
data_format=data_format,
name="conv1")
self.conv2 = InceptConv(in_channels=32,
out_channels=32,
kernel_size=3,
strides=1,
padding=0,
data_format=data_format,
name="conv2")
self.conv3 = InceptConv(in_channels=32,
out_channels=64,
kernel_size=3,
strides=1,
padding=1,
data_format=data_format,
name="conv3")
self.pool1 = MaxPool2d(pool_size=3,
strides=2,
padding=0,
data_format=data_format,
name="pool1")
self.conv4 = InceptConv(in_channels=64,
out_channels=80,
kernel_size=1,
strides=1,
padding=0,
data_format=data_format,
name="conv4")
self.conv5 = InceptConv(in_channels=80,
out_channels=192,
kernel_size=3,
strides=1,
padding=0,
data_format=data_format,
name="conv5")
self.pool2 = MaxPool2d(pool_size=3,
strides=2,
padding=0,
data_format=data_format,
name="pool2")
def __init__(self,
in_channels,
out_channels,
use_deptwise,
activation,
data_format="channels_last",
**kwargs):
super(HarDInitBlock, self).__init__(**kwargs)
mid_channels = out_channels // 2
self.conv1 = conv3x3_block(in_channels=in_channels,
out_channels=mid_channels,
strides=2,
activation=activation,
data_format=data_format,
name="conv1")
conv2_block_class = conv1x1_block if use_deptwise else conv3x3_block
self.conv2 = conv2_block_class(in_channels=mid_channels,
out_channels=out_channels,
activation=activation,
data_format=data_format,
name="conv2")
if use_deptwise:
self.downsample = dwconv3x3_block(in_channels=out_channels,
out_channels=out_channels,
strides=2,
activation=None,
data_format=data_format,
name="downsample")
else:
self.downsample = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="downsample")
def __init__(self, data_format="channels_last", **kwargs):
super(MaxPoolBranch, self).__init__(**kwargs)
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=0,
data_format=data_format,
name="pool")
def __init__(self,
channels,
residuals,
init_block_kernel_size,
init_block_channels,
in_channels=3,
in_size=(224, 224),
classes=1000,
data_format="channels_last",
**kwargs):
super(SqueezeNet, self).__init__(**kwargs)
self.in_size = in_size
self.classes = classes
self.data_format = data_format
self.features = tf.keras.Sequential(name="features")
self.features.add(
SqueezeInitBlock(in_channels=in_channels,
out_channels=init_block_channels,
kernel_size=init_block_kernel_size,
data_format=data_format,
name="init_block"))
in_channels = init_block_channels
for i, channels_per_stage in enumerate(channels):
stage = tf.keras.Sequential(name="stage{}".format(i + 1))
stage.add(
MaxPool2d(pool_size=3,
strides=2,
ceil_mode=True,
data_format=data_format,
name="pool{}".format(i + 1)))
for j, out_channels in enumerate(channels_per_stage):
expand_channels = out_channels // 2
squeeze_channels = out_channels // 8
stage.add(
FireUnit(in_channels=in_channels,
squeeze_channels=squeeze_channels,
expand1x1_channels=expand_channels,
expand3x3_channels=expand_channels,
residual=((residuals is not None)
and (residuals[i][j] == 1)),
data_format=data_format,
name="unit{}".format(j + 1)))
in_channels = out_channels
self.features.add(stage)
self.features.add(nn.Dropout(rate=0.5, name="dropout"))
self.output1 = tf.keras.Sequential(name="output1")
self.output1.add(
Conv2d(in_channels=in_channels,
out_channels=classes,
kernel_size=1,
data_format=data_format,
name="final_conv"))
self.output1.add(nn.ReLU())
self.output1.add(
nn.AveragePooling2D(pool_size=13,
strides=1,
data_format=data_format,
name="final_pool"))
def __init__(self,
in_channels,
out_channels,
kernel_size,
padding,
data_format="channels_last",
**kwargs):
super(DPNInitBlock, self).__init__(**kwargs)
self.conv = Conv2d(in_channels=in_channels,
out_channels=out_channels,
kernel_size=kernel_size,
strides=2,
padding=padding,
use_bias=False,
data_format=data_format,
name="conv")
self.bn = dpn_batch_norm(channels=out_channels,
data_format=data_format,
name="bn")
self.activ = nn.ReLU()
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(StemBlock, self).__init__(**kwargs)
mid1_channels = out_channels // 2
mid2_channels = out_channels * 2
self.first_conv = conv3x3_block(in_channels=in_channels,
out_channels=out_channels,
strides=2,
data_format=data_format,
name="first_conv")
self.branches = Concurrent(data_format=data_format, name="branches")
self.branches.add(
PeleeBranch1(in_channels=out_channels,
out_channels=out_channels,
mid_channels=mid1_channels,
strides=2,
data_format=data_format,
name="branch1"))
self.branches.add(
MaxPool2d(pool_size=2,
strides=2,
padding=0,
data_format=data_format,
name="branch2"))
self.last_conv = conv1x1_block(in_channels=mid2_channels,
out_channels=out_channels,
data_format=data_format,
name="last_conv")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(SEInitBlock, self).__init__(**kwargs)
mid_channels = out_channels // 2
self.conv1 = conv3x3_block(in_channels=in_channels,
out_channels=mid_channels,
strides=2,
data_format=data_format,
name="conv1")
self.conv2 = conv3x3_block(in_channels=mid_channels,
out_channels=mid_channels,
data_format=data_format,
name="conv2")
self.conv3 = conv3x3_block(in_channels=mid_channels,
out_channels=out_channels,
data_format=data_format,
name="conv3")
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
in_channels,
out_channels,
avg_pool,
use_bias,
use_bn,
data_format="channels_last",
**kwargs):
super(InceptionPoolBranch, self).__init__(**kwargs)
if avg_pool:
self.pool = AvgPool2d(
pool_size=3,
strides=1,
padding=1,
ceil_mode=True,
# count_include_pad=True,
data_format=data_format,
name="pool")
else:
self.pool = MaxPool2d(pool_size=3,
strides=1,
padding=1,
ceil_mode=True,
data_format=data_format,
name="pool")
self.conv = conv1x1_block(in_channels=in_channels,
out_channels=out_channels,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="conv")
def __init__(self,
in_channels_list,
out_channels_list,
links_list,
use_deptwise,
use_dropout,
downsampling,
activation,
data_format="channels_last",
**kwargs):
super(HarDUnit, self).__init__(**kwargs)
self.data_format = data_format
self.links_list = links_list
self.use_dropout = use_dropout
self.downsampling = downsampling
self.blocks = SimpleSequential(name="blocks")
for i in range(len(links_list)):
in_channels = in_channels_list[i]
out_channels = out_channels_list[i]
if use_deptwise:
unit = invdwsconv3x3_block(in_channels=in_channels,
out_channels=out_channels,
pw_activation=activation,
dw_activation=None,
data_format=data_format,
name="block{}".format(i + 1))
else:
unit = conv3x3_block(in_channels=in_channels,
out_channels=out_channels,
data_format=data_format,
name="block{}".format(i + 1))
self.blocks.add(unit)
if self.use_dropout:
self.dropout = nn.Dropout(rate=0.1, name="dropout")
self.conv = conv1x1_block(in_channels=in_channels_list[-1],
out_channels=out_channels_list[-1],
activation=activation,
data_format=data_format,
name="conv")
if self.downsampling:
if use_deptwise:
self.downsample = dwconv3x3_block(
in_channels=out_channels_list[-1],
out_channels=out_channels_list[-1],
strides=2,
activation=None,
data_format=data_format,
name="downsample")
else:
self.downsample = MaxPool2d(pool_size=2,
strides=2,
data_format=data_format,
name="downsample")
def __init__(self,
channels,
odd_pointwise,
avg_pool_size,
cls_activ,
alpha=0.1,
in_channels=3,
in_size=(224, 224),
classes=1000,
data_format="channels_last",
**kwargs):
super(DarkNet, self).__init__(**kwargs)
self.in_size = in_size
self.classes = classes
self.data_format = data_format
self.features = tf.keras.Sequential(name="features")
for i, channels_per_stage in enumerate(channels):
stage = tf.keras.Sequential(name="stage{}".format(i + 1))
for j, out_channels in enumerate(channels_per_stage):
stage.add(
dark_convYxY(in_channels=in_channels,
out_channels=out_channels,
alpha=alpha,
pointwise=(len(channels_per_stage) > 1)
and not (((j + 1) % 2 == 1) ^ odd_pointwise),
data_format=data_format,
name="unit{}".format(j + 1)))
in_channels = out_channels
if i != len(channels) - 1:
stage.add(
MaxPool2d(pool_size=2,
strides=2,
data_format=data_format,
name="pool{}".format(i + 1)))
self.features.add(stage)
self.output1 = tf.keras.Sequential(name="output1")
self.output1.add(
Conv2d(in_channels=in_channels,
out_channels=classes,
kernel_size=1,
data_format=data_format,
name="final_conv"))
if cls_activ:
self.output1.add(nn.LeakyReLU(alpha=alpha))
self.output1.add(
nn.AveragePooling2D(pool_size=avg_pool_size,
strides=1,
data_format=data_format,
name="final_pool"))
def __init__(self,
in_channels,
out_channels,
activation,
data_format="channels_last",
**kwargs):
super(IbpDownBlock, self).__init__(**kwargs)
self.down = MaxPool2d(pool_size=2,
strides=2,
data_format=data_format,
name="down")
self.res = IbpResUnit(in_channels=in_channels,
out_channels=out_channels,
activation=activation,
data_format=data_format,
name="res")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(IBNbResInitBlock, self).__init__(**kwargs)
self.conv = ibnb_conv7x7_block(in_channels=in_channels,
out_channels=out_channels,
strides=2,
data_format=data_format,
name="conv")
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
channels,
init_block_channels,
in_channels=3,
in_size=(224, 224),
classes=1000,
data_format="channels_last",
**kwargs):
super(DiracNetV2, self).__init__(**kwargs)
self.in_size = in_size
self.classes = classes
self.data_format = data_format
self.features = tf.keras.Sequential(name="features")
self.features.add(
DiracInitBlock(in_channels=in_channels,
out_channels=init_block_channels,
data_format=data_format,
name="init_block"))
in_channels = init_block_channels
for i, channels_per_stage in enumerate(channels):
stage = tf.keras.Sequential(name="stage{}".format(i + 1))
for j, out_channels in enumerate(channels_per_stage):
stage.add(
dirac_conv3x3(in_channels=in_channels,
out_channels=out_channels,
data_format=data_format,
name="unit{}".format(j + 1)))
in_channels = out_channels
if i != len(channels) - 1:
stage.add(
MaxPool2d(pool_size=2,
strides=2,
padding=0,
data_format=data_format,
name="pool{}".format(i + 1)))
self.features.add(stage)
self.features.add(nn.ReLU(name="final_activ"))
self.features.add(
nn.AveragePooling2D(pool_size=7,
strides=1,
data_format=data_format,
name="final_pool"))
self.output1 = nn.Dense(units=classes,
input_dim=in_channels,
name="output1")
def __init__(self,
in_channels,
out_channels,
strides,
reps,
start_with_relu=True,
grow_first=True,
data_format="channels_last",
**kwargs):
super(XceptionUnit, self).__init__(**kwargs)
self.resize_identity = (in_channels != out_channels) or (strides != 1)
if self.resize_identity:
self.identity_conv = conv1x1_block(in_channels=in_channels,
out_channels=out_channels,
strides=strides,
activation=None,
data_format=data_format,
name="identity_conv")
self.body = SimpleSequential(name="body")
for i in range(reps):
if (grow_first and (i == 0)) or ((not grow_first) and
(i == reps - 1)):
in_channels_i = in_channels
out_channels_i = out_channels
else:
if grow_first:
in_channels_i = out_channels
out_channels_i = out_channels
else:
in_channels_i = in_channels
out_channels_i = in_channels
activate = start_with_relu if (i == 0) else True
self.body.children.append(
dws_conv3x3_block(in_channels=in_channels_i,
out_channels=out_channels_i,
activate=activate,
data_format=data_format,
name="block{}".format(i + 1)))
if strides != 1:
self.body.children.append(
MaxPool2d(pool_size=3,
strides=strides,
padding=1,
data_format=data_format,
name="pool"))
def __init__(self,
strides=2,
extra_padding=False,
data_format="channels_last",
**kwargs):
super(PnasMaxPoolBlock, self).__init__(**kwargs)
self.extra_padding = extra_padding
self.data_format = data_format
self.pool = MaxPool2d(pool_size=3,
strides=strides,
padding=1,
data_format=data_format,
name="pool")
if self.extra_padding:
self.pad = nn.ZeroPadding2D(padding=((1, 0), (1, 0)),
data_format=data_format)
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(ShuffleInitBlock, self).__init__(**kwargs)
self.conv = conv3x3_block(in_channels=in_channels,
out_channels=out_channels,
strides=2,
data_format=data_format,
name="conv")
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=0,
ceil_mode=True,
data_format=data_format,
name="pool")
def __init__(self,
channels,
kernel_sizes,
strides,
paddings,
use_lrn,
in_channels=3,
in_size=(224, 224),
classes=1000,
data_format="channels_last",
**kwargs):
super(AlexNet, self).__init__(**kwargs)
self.in_size = in_size
self.classes = classes
self.data_format = data_format
self.features = tf.keras.Sequential(name="features")
for i, channels_per_stage in enumerate(channels):
use_lrn_i = use_lrn and (i in [0, 1])
stage = tf.keras.Sequential(name="stage{}".format(i + 1))
for j, out_channels in enumerate(channels_per_stage):
stage.add(AlexConv(
in_channels=in_channels,
out_channels=out_channels,
kernel_size=kernel_sizes[i][j],
strides=strides[i][j],
padding=paddings[i][j],
use_lrn=use_lrn_i,
data_format=data_format,
name="unit{}".format(j + 1)))
in_channels = out_channels
stage.add(MaxPool2d(
pool_size=3,
strides=2,
padding=0,
ceil_mode=True,
data_format=data_format,
name="pool{}".format(i + 1)))
self.features.add(stage)
in_channels = in_channels * 6 * 6
self.output1 = AlexOutputBlock(
in_channels=in_channels,
classes=classes,
name="output1")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(DiracInitBlock, self).__init__(**kwargs)
self.conv = Conv2d(in_channels=in_channels,
out_channels=out_channels,
kernel_size=7,
strides=2,
padding=3,
use_bias=True,
data_format=data_format,
name="conv")
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
in_channels,
out_channels,
activation,
data_format="channels_last",
**kwargs):
super(IbpBackbone, self).__init__(**kwargs)
self.data_format = data_format
dilations = (3, 3, 4, 4, 5, 5)
mid1_channels = out_channels // 4
mid2_channels = out_channels // 2
self.conv1 = conv7x7_block(in_channels=in_channels,
out_channels=mid1_channels,
strides=2,
activation=activation,
data_format=data_format,
name="conv1")
self.res1 = IbpResUnit(in_channels=mid1_channels,
out_channels=mid2_channels,
activation=activation,
data_format=data_format,
name="res1")
self.pool = MaxPool2d(pool_size=2,
strides=2,
data_format=data_format,
name="pool")
self.res2 = IbpResUnit(in_channels=mid2_channels,
out_channels=mid2_channels,
activation=activation,
data_format=data_format,
name="res2")
self.dilation_branch = SimpleSequential(name="dilation_branch")
for i, dilation in enumerate(dilations):
self.dilation_branch.add(
conv3x3_block(in_channels=mid2_channels,
out_channels=mid2_channels,
padding=dilation,
dilation=dilation,
activation=activation,
data_format=data_format,
name="block{}".format(i + 1)))
def __init__(self,
in_channels,
out_channels,
do_nms,
data_format="channels_last",
**kwargs):
super(CenterNetHeatmapBlock, self).__init__(**kwargs)
self.do_nms = do_nms
self.head = CenterNetHeadBlock(in_channels=in_channels,
out_channels=out_channels,
data_format=data_format,
name="head")
self.sigmoid = tf.nn.sigmoid
if self.do_nms:
self.pool = MaxPool2d(pool_size=3,
strides=1,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
in_channels,
out_channels,
data_format="channels_last",
**kwargs):
super(ShuffleInitBlock, self).__init__(**kwargs)
self.conv = conv3x3(in_channels=in_channels,
out_channels=out_channels,
strides=2,
data_format=data_format,
name="conv")
self.bn = BatchNorm(
# in_channels=out_channels,
data_format=data_format,
name="bn")
self.activ = nn.ReLU()
self.pool = MaxPool2d(pool_size=3,
strides=2,
padding=1,
data_format=data_format,
name="pool")
def __init__(self,
in_channels,
mid1_channels_list,
mid2_channels_list,
use_bias,
use_bn,
data_format="channels_last",
**kwargs):
super(ReductionBlock, self).__init__(**kwargs)
assert (len(mid1_channels_list) == 2)
assert (len(mid2_channels_list) == 4)
self.branches = Concurrent(data_format=data_format, name="branches")
self.branches.children.append(
Inception3x3Branch(in_channels=in_channels,
out_channels=mid2_channels_list[1],
mid_channels=mid1_channels_list[0],
strides=2,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="branch1"))
self.branches.children.append(
InceptionDouble3x3Branch(in_channels=in_channels,
out_channels=mid2_channels_list[2],
mid_channels=mid1_channels_list[1],
strides=2,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="branch2"))
self.branches.children.append(
MaxPool2d(pool_size=3,
strides=2,
padding=0,
ceil_mode=True,
data_format=data_format,
name="branch3"))
def __init__(self,
in_channels,
out_channels,
branch_channels,
num_branches,
resize,
use_residual,
data_format="channels_last",
**kwargs):
super(VoVUnit, self).__init__(**kwargs)
self.resize = resize
self.use_residual = use_residual
if self.resize:
self.pool = MaxPool2d(
pool_size=3,
strides=2,
ceil_mode=True,
data_format=data_format,
name="pool")
self.branches = SequentialConcurrent(
data_format=data_format,
name="branches")
branch_in_channels = in_channels
for i in range(num_branches):
self.branches.add(conv3x3_block(
in_channels=branch_in_channels,
out_channels=branch_channels,
data_format=data_format,
name="branch{}".format(i + 1)))
branch_in_channels = branch_channels
self.concat_conv = conv1x1_block(
in_channels=(in_channels + num_branches * branch_channels),
out_channels=out_channels,
data_format=data_format,
name="concat_conv")
def __init__(self,
channels,
use_bias=True,
use_bn=False,
in_channels=3,
in_size=(224, 224),
classes=1000,
data_format="channels_last",
**kwargs):
super(VGG, self).__init__(**kwargs)
self.in_size = in_size
self.classes = classes
self.data_format = data_format
self.features = tf.keras.Sequential(name="features")
for i, channels_per_stage in enumerate(channels):
stage = tf.keras.Sequential(name="stage{}".format(i + 1))
for j, out_channels in enumerate(channels_per_stage):
stage.add(conv3x3_block(
in_channels=in_channels,
out_channels=out_channels,
use_bias=use_bias,
use_bn=use_bn,
data_format=data_format,
name="unit{}".format(j + 1)))
in_channels = out_channels
stage.add(MaxPool2d(
pool_size=2,
strides=2,
padding=0,
data_format=data_format,
name="pool{}".format(i + 1)))
self.features.add(stage)
self.output1 = VGGOutputBlock(
in_channels=(in_channels * 7 * 7),
classes=classes,
name="output1")
