def _make_transition_layer( self, num_channels_pre_layer, num_channels_cur_layer): num_branches_cur = len(num_channels_cur_layer) num_branches_pre = len(num_channels_pre_layer) transition_layers = [] for i in range(num_branches_cur): if i < num_branches_pre: if num_channels_cur_layer[i] != num_channels_pre_layer[i]: transition_layers.append(nn.Sequential( nn.Conv2d(num_channels_pre_layer[i], num_channels_cur_layer[i], 3, 1, 1, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)( num_channels_cur_layer[i]), nn.ReLU(inplace=True))) else: transition_layers.append(None) else: conv3x3s = [] for j in range(i+1-num_branches_pre): inchannels = num_channels_pre_layer[-1] outchannels = num_channels_cur_layer[i] \ if j == i-num_branches_pre else inchannels conv3x3s.append(nn.Sequential( nn.Conv2d( inchannels, outchannels, 3, 2, 1, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(outchannels), nn.ReLU(inplace=True))) transition_layers.append(nn.Sequential(*conv3x3s)) return nn.ModuleList(transition_layers)
def __init__(self, input_num, num1, num2, dilation_rate, drop_out, norm_type): super(_DenseAsppBlock, self).__init__() self.add_module( 'conv1', nn.Conv2d(in_channels=input_num, out_channels=num1, kernel_size=1)), self.add_module( 'norm1', ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_features=num1)), self.add_module('relu1', nn.ReLU(inplace=False)), self.add_module( 'conv2', nn.Conv2d(in_channels=num1, out_channels=num2, kernel_size=3, dilation=dilation_rate, padding=dilation_rate)), self.add_module( 'norm2', ModuleHelper.BatchNorm2d(norm_type=norm_type)( num_features=input_num)), self.add_module('relu2', nn.ReLU(inplace=False)), self.drop_rate = drop_out
def __init__(self, num_input_features, growth_rate, bn_size, drop_rate, norm_type): super(_DenseLayer, self).__init__() self.add_module( 'norm1', ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_input_features)), self.add_module('relu1', nn.ReLU(inplace=True)), self.add_module( 'conv1', nn.Conv2d(num_input_features, bn_size * growth_rate, kernel_size=1, stride=1, bias=False)), self.add_module( 'norm2', ModuleHelper.BatchNorm2d(norm_type=norm_type)(bn_size * growth_rate)), self.add_module('relu2', nn.ReLU(inplace=True)), self.add_module( 'conv2', nn.Conv2d(bn_size * growth_rate, growth_rate, kernel_size=3, stride=1, padding=1, bias=False)), self.drop_rate = drop_rate
def __init__(self, inplanes, planes, stride=1, downsample=None): super(BasicBlock, self).__init__() self.conv1 = conv3x3(inplanes, planes, stride) self.bn1 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes) self.relu = nn.ReLU(inplace=True) self.conv2 = conv3x3(planes, planes) self.bn2 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes) self.downsample = downsample self.stride = stride
def __init__(self, **kwargs): super(HighResolutionNet, self).__init__() self.num_features = 720 # stem net self.conv1 = nn.Conv2d(3, 64, kernel_size=3, stride=2, padding=1, bias=False) self.bn1 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(64) self.conv2 = nn.Conv2d(64, 64, kernel_size=3, stride=2, padding=1, bias=False) self.bn2 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(64) self.relu = nn.ReLU(inplace=True) self.stage1_cfg = {'NUM_MODULES':1, 'NUM_BRANCHES':1, 'BLOCK':'BOTTLENECK', 'NUM_BLOCKS':[4], 'NUM_CHANNELS':[64], 'FUSE_METHOD':'SUM'} num_channels = self.stage1_cfg['NUM_CHANNELS'][0] block = blocks_dict[self.stage1_cfg['BLOCK']] num_blocks = self.stage1_cfg['NUM_BLOCKS'][0] self.layer1 = self._make_layer(block, 64, num_channels, num_blocks) stage1_out_channel = block.expansion*num_channels self.stage2_cfg = {'NUM_MODULES':1, 'NUM_BRANCHES':2, 'BLOCK':'BASIC', 'NUM_BLOCKS':[4,4], 'NUM_CHANNELS':[48,96], 'FUSE_METHOD':'SUM'} num_channels = self.stage2_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage2_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition1 = self._make_transition_layer( [stage1_out_channel], num_channels) self.stage2, pre_stage_channels = self._make_stage( self.stage2_cfg, num_channels) self.stage3_cfg = {'NUM_MODULES':4, 'NUM_BRANCHES':3, 'BLOCK':'BASIC', 'NUM_BLOCKS':[4,4,4], 'NUM_CHANNELS':[48,96,192], 'FUSE_METHOD':'SUM'} num_channels = self.stage3_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage3_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition2 = self._make_transition_layer( pre_stage_channels, num_channels) self.stage3, pre_stage_channels = self._make_stage( self.stage3_cfg, num_channels) self.stage4_cfg = {'NUM_MODULES':3, 'NUM_BRANCHES':4, 'BLOCK':'BASIC', 'NUM_BLOCKS':[4,4,4,4], 'NUM_CHANNELS':[48,96,192,384], 'FUSE_METHOD':'SUM'} num_channels = self.stage4_cfg['NUM_CHANNELS'] block = blocks_dict[self.stage4_cfg['BLOCK']] num_channels = [ num_channels[i] * block.expansion for i in range(len(num_channels))] self.transition3 = self._make_transition_layer( pre_stage_channels, num_channels) self.stage4, pre_stage_channels = self._make_stage( self.stage4_cfg, num_channels, multi_scale_output=True) last_inp_channels = np.int(np.sum(pre_stage_channels))
def __init__(self, inplanes, planes, stride=1, downsample=None): super(Bottleneck, self).__init__() self.conv1 = nn.Conv2d(inplanes, planes, kernel_size=1, bias=False) self.bn1 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes) self.conv2 = nn.Conv2d(planes, planes, kernel_size=3, stride=stride, padding=1, bias=False) self.bn2 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes) self.conv3 = nn.Conv2d(planes, planes * self.expansion, kernel_size=1, bias=False) self.bn3 = ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes * self.expansion) self.relu = nn.ReLU(inplace=True) self.downsample = downsample self.stride = stride
def __init__(self, input_nc, ndf=64, n_layers=3, norm_type=None): """Construct a PatchGAN discriminator Parameters: input_nc (int) -- the number of channels in input images ndf (int) -- the number of filters in the last conv layer n_layers (int) -- the number of conv layers in the discriminator norm_layer -- normalization layer """ super(NLayerDiscriminator, self).__init__() use_bias = (norm_type == 'instancenorm') kw = 4 padw = 1 sequence = [ nn.Conv2d(input_nc, ndf, kernel_size=kw, stride=2, padding=padw), nn.LeakyReLU(0.2, True) ] nf_mult = 1 nf_mult_prev = 1 for n in range(1, n_layers): # gradually increase the number of filters nf_mult_prev = nf_mult nf_mult = min(2**n, 8) sequence += [ nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=2, padding=padw, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(ndf * nf_mult), nn.LeakyReLU(0.2, True) ] nf_mult_prev = nf_mult nf_mult = min(2**n_layers, 8) sequence += [ nn.Conv2d(ndf * nf_mult_prev, ndf * nf_mult, kernel_size=kw, stride=1, padding=padw, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(ndf * nf_mult), nn.LeakyReLU(0.2, True) ] sequence += [ nn.Conv2d(ndf * nf_mult, 1, kernel_size=kw, stride=1, padding=padw) ] # output 1 channel prediction map self.model = nn.Sequential(*sequence)
def __init__(self, input_nc, ndf=64, norm_type=None): """Construct a 1x1 PatchGAN discriminator Parameters: input_nc (int) -- the number of channels in input images ndf (int) -- the number of filters in the last conv layer norm_layer -- normalization layer """ super(PixelDiscriminator, self).__init__() use_bias = (norm_type == 'instancenorm') self.net = [ nn.Conv2d(input_nc, ndf, kernel_size=1, stride=1, padding=0), nn.LeakyReLU(0.2, True), nn.Conv2d(ndf, ndf * 2, kernel_size=1, stride=1, padding=0, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(ndf * 2), nn.LeakyReLU(0.2, True), nn.Conv2d(ndf * 2, 1, kernel_size=1, stride=1, padding=0, bias=use_bias) ] self.net = nn.Sequential(*self.net)
def _make_layer(self, block, planes, blocks, stride=1, norm_type=None): downsample = None if stride != 1 or self.inplanes != planes * block.expansion: downsample = nn.Sequential( nn.Conv2d(self.inplanes, planes * block.expansion, kernel_size=1, stride=stride, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(planes * block.expansion), ) layers = [] layers.append( block(self.inplanes, planes, stride, downsample, norm_type=norm_type)) self.inplanes = planes * block.expansion for i in range(1, blocks): layers.append(block(self.inplanes, planes, norm_type=norm_type)) return nn.Sequential(*layers)
def build_conv_block(self, dim, padding_type, norm_type, use_dropout, use_bias): """Construct a convolutional block. Parameters: dim (int) -- the number of channels in the conv layer. padding_type (str) -- the name of padding layer: reflect | replicate | zero norm_layer -- normalization layer use_dropout (bool) -- if use dropout layers. use_bias (bool) -- if the conv layer uses bias or not Returns a conv block (with a conv layer, a normalization layer, and a non-linearity layer (ReLU)) """ conv_block = [] p = 0 if padding_type == 'reflect': conv_block += [nn.ReflectionPad2d(1)] elif padding_type == 'replicate': conv_block += [nn.ReplicationPad2d(1)] elif padding_type == 'zero': p = 1 else: raise NotImplementedError('padding [%s] is not implemented' % padding_type) conv_block += [ nn.Conv2d(dim, dim, kernel_size=3, padding=p, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(dim), nn.ReLU(True) ] if use_dropout: conv_block += [nn.Dropout(0.5)] p = 0 if padding_type == 'reflect': conv_block += [nn.ReflectionPad2d(1)] elif padding_type == 'replicate': conv_block += [nn.ReplicationPad2d(1)] elif padding_type == 'zero': p = 1 else: raise NotImplementedError('padding [%s] is not implemented' % padding_type) conv_block += [ nn.Conv2d(dim, dim, kernel_size=3, padding=p, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(dim) ] return nn.Sequential(*conv_block)
def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16), num_init_features=64, bn_size=4, drop_rate=0, num_classes=1000, norm_type=None): super(DenseNet, self).__init__() # First convolution self.features = nn.Sequential(OrderedDict([ ('conv0', nn.Conv2d(3, num_init_features, kernel_size=7, stride=2, padding=3, bias=False)), ('norm0', ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_init_features)), ('relu0', nn.ReLU(inplace=True)), ('pool0', nn.MaxPool2d(kernel_size=3, stride=2, padding=1)), ])) # Each denseblock num_features = num_init_features for i, num_layers in enumerate(block_config): block = _DenseBlock(num_layers=num_layers, num_input_features=num_features, bn_size=bn_size, growth_rate=growth_rate, drop_rate=drop_rate, norm_type=norm_type) self.features.add_module('denseblock%d' % (i + 1), block) num_features = num_features + num_layers * growth_rate if i != len(block_config) - 1: trans = _Transition(num_input_features=num_features, num_output_features=num_features // 2, norm_type=norm_type) avg_pool = nn.AvgPool2d(kernel_size=2, stride=2) self.features.add_module('transition%d' % (i + 1), trans) self.features.add_module('transition%s_pool' % (i + 1), avg_pool) num_features = num_features // 2 self.num_features = num_features # Final batch norm self.features.add_module('norm5', ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_features)) # Linear layer self.classifier = nn.Linear(num_features, num_classes) # Official init from torch repo. for m in self.modules(): if isinstance(m, nn.Conv2d): nn.init.kaiming_normal_(m.weight) elif isinstance(m, ModuleHelper.BatchNorm2d(norm_type=norm_type, ret_cls=True)): nn.init.constant_(m.weight, 1) nn.init.constant_(m.bias, 0) elif isinstance(m, nn.Linear): nn.init.constant_(m.bias, 0)
def _make_fuse_layers(self): if self.num_branches == 1: return None num_branches = self.num_branches num_inchannels = self.num_inchannels fuse_layers = [] for i in range(num_branches if self.multi_scale_output else 1): fuse_layer = [] for j in range(num_branches): if j > i: fuse_layer.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_inchannels[i], 1, 1, 0, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_inchannels[i]))) elif j == i: fuse_layer.append(None) else: conv3x3s = [] for k in range(i-j): if k == i - j - 1: num_outchannels_conv3x3 = num_inchannels[i] conv3x3s.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_outchannels_conv3x3))) else: num_outchannels_conv3x3 = num_inchannels[j] conv3x3s.append(nn.Sequential( nn.Conv2d(num_inchannels[j], num_outchannels_conv3x3, 3, 2, 1, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_outchannels_conv3x3), nn.ReLU(inplace=True))) fuse_layer.append(nn.Sequential(*conv3x3s)) fuse_layers.append(nn.ModuleList(fuse_layer)) return nn.ModuleList(fuse_layers)
def freeze_bn(net, norm_type=None): for m in net.modules(): if isinstance(m, nn.BatchNorm2d) or isinstance(m, nn.BatchNorm1d) or isinstance(m, nn.BatchNorm3d): m.eval() if norm_type is not None: from model.tools.module_helper import ModuleHelper if isinstance(m, ModuleHelper.BatchNorm2d(norm_type=norm_type, ret_cls=True)) \ or isinstance(m, ModuleHelper.BatchNorm1d(norm_type=norm_type, ret_cls=True)) \ or isinstance(m, ModuleHelper.BatchNorm3d(norm_type=norm_type, ret_cls=True)): m.eval()
def __init__(self, num_input_features, num_output_features, norm_type): super(_Transition, self).__init__() self.add_module( 'norm', ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_input_features)) self.add_module('relu', nn.ReLU(inplace=True)) self.add_module( 'conv', nn.Conv2d(num_input_features, num_output_features, kernel_size=1, stride=1, bias=False))
def __init__(self, low_in_channels, high_in_channels, out_channels, key_channels, value_channels, dropout, sizes=([1]), norm_type=None,psp_size=(1,3,6,8)): super(AFNB, self).__init__() self.stages = [] self.norm_type = norm_type self.psp_size=psp_size self.stages = nn.ModuleList( [self._make_stage([low_in_channels, high_in_channels], out_channels, key_channels, value_channels, size) for size in sizes]) self.conv_bn_dropout = nn.Sequential( nn.Conv2d(out_channels + high_in_channels, out_channels, kernel_size=1, padding=0), ModuleHelper.BatchNorm2d(norm_type=self.norm_type)(out_channels), nn.Dropout2d(dropout) )
def __init__(self, in_channel, out_channel, kernel_size, stride=1, padding=0, dilation=1, groups=1, norm_type=None): super(ConvBn, self).__init__() self.conv_bn = nn.Sequential( nn.Conv2d(in_channel, out_channel, kernel_size, stride, padding, dilation, groups, False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(out_channel))
def _make_one_branch(self, branch_index, block, num_blocks, num_channels, stride=1): downsample = None if stride != 1 or \ self.num_inchannels[branch_index] != num_channels[branch_index] * block.expansion: downsample = nn.Sequential( nn.Conv2d(self.num_inchannels[branch_index], num_channels[branch_index] * block.expansion, kernel_size=1, stride=stride, bias=False), ModuleHelper.BatchNorm2d(norm_type=norm_type)(num_channels[branch_index] * block.expansion), ) layers = [] layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index], stride, downsample)) self.num_inchannels[branch_index] = \ num_channels[branch_index] * block.expansion for i in range(1, num_blocks[branch_index]): layers.append(block(self.num_inchannels[branch_index], num_channels[branch_index])) return nn.Sequential(*layers)
def __init__(self, block, layers, num_classes=1000, deep_base=False, norm_type=None): super(ResNet, self).__init__() self.inplanes = 128 if deep_base else 64 if deep_base: self.prefix = nn.Sequential( OrderedDict([ ('conv1', nn.Conv2d(3, 64, kernel_size=3, stride=2, padding=1, bias=False)), ('bn1', ModuleHelper.BatchNorm2d(norm_type=norm_type)(64)), ('relu1', nn.ReLU(inplace=False)), ('conv2', nn.Conv2d(64, 64, kernel_size=3, stride=1, padding=1, bias=False)), ('bn2', ModuleHelper.BatchNorm2d(norm_type=norm_type)(64)), ('relu2', nn.ReLU(inplace=False)), ('conv3', nn.Conv2d(64, 128, kernel_size=3, stride=1, padding=1, bias=False)), ('bn3', ModuleHelper.BatchNorm2d(norm_type=norm_type)( self.inplanes)), ('relu3', nn.ReLU(inplace=False)) ])) else: self.prefix = nn.Sequential( OrderedDict([('conv1', nn.Conv2d(3, 64, kernel_size=7, stride=2, padding=3, bias=False)), ('bn1', ModuleHelper.BatchNorm2d(norm_type=norm_type)( self.inplanes)), ('relu', nn.ReLU(inplace=False))])) self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1, ceil_mode=True) # change. self.layer1 = self._make_layer(block, 64, layers[0], norm_type=norm_type) self.layer2 = self._make_layer(block, 128, layers[1], stride=2, norm_type=norm_type) self.layer3 = self._make_layer(block, 256, layers[2], stride=2, norm_type=norm_type) self.layer4 = self._make_layer(block, 512, layers[3], stride=2, norm_type=norm_type) self.avgpool = nn.AvgPool2d(7, stride=1) self.fc = nn.Linear(512 * block.expansion, num_classes) for m in self.modules(): if isinstance(m, nn.Conv2d): n = m.kernel_size[0] * m.kernel_size[1] * m.out_channels m.weight.data.normal_(0, math.sqrt(2. / n)) elif isinstance( m, ModuleHelper.BatchNorm2d(norm_type=norm_type, ret_cls=True)): m.weight.data.fill_(1) m.bias.data.zero_()
def __init__(self, input_nc, output_nc, ngf=64, norm_type=None, use_dropout=False, n_blocks=6, padding_type='reflect'): """Construct a Resnet-based generator Parameters: input_nc (int) -- the number of channels in input images output_nc (int) -- the number of channels in output images ngf (int) -- the number of filters in the last conv layer norm_layer -- normalization layer use_dropout (bool) -- if use dropout layers n_blocks (int) -- the number of ResNet blocks padding_type (str) -- the name of padding layer in conv layers: reflect | replicate | zero """ assert (n_blocks >= 0) super(ResNetGenerator, self).__init__() use_bias = (norm_type == 'instancenorm') model = [ nn.ReflectionPad2d(3), nn.Conv2d(input_nc, ngf, kernel_size=7, padding=0, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(ngf), nn.ReLU(True) ] n_downsampling = 2 for i in range(n_downsampling): # add downsampling layers mult = 2**i model += [ nn.Conv2d(ngf * mult, ngf * mult * 2, kernel_size=3, stride=2, padding=1, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(ngf * mult * 2), nn.ReLU(True) ] mult = 2**n_downsampling for i in range(n_blocks): # add ResNet blocks model += [ ResnetBlock(ngf * mult, padding_type=padding_type, norm_type=norm_type, use_dropout=use_dropout, use_bias=use_bias) ] for i in range(n_downsampling): # add upsampling layers mult = 2**(n_downsampling - i) model += [ nn.ConvTranspose2d(ngf * mult, int(ngf * mult / 2), kernel_size=3, stride=2, padding=1, output_padding=1, bias=use_bias), ModuleHelper.BatchNorm2d(norm_type=norm_type)(int(ngf * mult / 2)), nn.ReLU(True) ] model += [nn.ReflectionPad2d(3)] model += [nn.Conv2d(ngf, output_nc, kernel_size=7, padding=0)] model += [nn.Tanh()] self.model = nn.Sequential(*model)
def __init__(self, outer_nc, inner_nc, input_nc=None, submodule=None, outermost=False, innermost=False, norm_type=None, use_dropout=False): """Construct a Unet submodule with skip connections. Parameters: outer_nc (int) -- the number of filters in the outer conv layer inner_nc (int) -- the number of filters in the inner conv layer input_nc (int) -- the number of channels in input images/features submodule (UnetSkipConnectionBlock) -- previously defined submodules outermost (bool) -- if this module is the outermost module innermost (bool) -- if this module is the innermost module norm_layer -- normalization layer user_dropout (bool) -- if use dropout layers. """ super(UnetSkipConnectionBlock, self).__init__() self.outermost = outermost use_bias = (norm_type == 'instancenorm') if input_nc is None: input_nc = outer_nc downconv = nn.Conv2d(input_nc, inner_nc, kernel_size=4, stride=2, padding=1, bias=use_bias) downrelu = nn.LeakyReLU(0.2, True) downnorm = ModuleHelper.BatchNorm2d(norm_type=norm_type)(inner_nc) uprelu = nn.ReLU(True) upnorm = ModuleHelper.BatchNorm2d(norm_type=norm_type)(outer_nc) if outermost: upconv = nn.ConvTranspose2d(inner_nc * 2, outer_nc, kernel_size=4, stride=2, padding=1) down = [downconv] up = [uprelu, upconv, nn.Tanh()] model = down + [submodule] + up elif innermost: upconv = nn.ConvTranspose2d(inner_nc, outer_nc, kernel_size=4, stride=2, padding=1, bias=use_bias) down = [downrelu, downconv] up = [uprelu, upconv, upnorm] model = down + up else: upconv = nn.ConvTranspose2d(inner_nc * 2, outer_nc, kernel_size=4, stride=2, padding=1, bias=use_bias) down = [downrelu, downconv, downnorm] up = [uprelu, upconv, upnorm] if use_dropout: model = down + [submodule] + up + [nn.Dropout(0.5)] else: model = down + [submodule] + up self.model = nn.Sequential(*model)