models/hrnet125.py

import torch
import torch.nn
import torch.functional
import torch.nn.functional


class hrnet125(torch.nn.Module):
    def __init__(self):
        super().__init__()

        self.conv1 = torch.nn.modules.conv.Conv2d(in_channels=3, out_channels=64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.bn1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.relu1 = torch.nn.modules.activation.ReLU(False)
        self.downsample_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.downsample_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.downsample_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.downsample_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_relu1 = torch.nn.modules.activation.ReLU(False)
        self.downsample_conv2_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(1, 1), padding=(3, 3), dilation=(3, 3), groups=64, bias=True, padding_mode="zeros")
        self.downsample_conv2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_conv2_2 = torch.nn.modules.activation.ReLU(False)
        self.downsample_conv2_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.downsample_conv2_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_downsample_res_conv_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.downsample_downsample_res_conv_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_downsample_res_conv_2 = torch.nn.modules.activation.ReLU(False)
        self.downsample_downsample_res_conv_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.downsample_downsample_res_conv_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.downsample_relu2 = torch.nn.modules.activation.ReLU(False)
        self.layer1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.layer1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.layer1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.layer1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.layer1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.layer1_1_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.layer1_1_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.layer1_1_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.layer1_1_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.layer1_1_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.cut1_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.cut1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.cut1_2 = torch.nn.modules.activation.ReLU(False)
        self.transition1_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.transition1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition1_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.transition1_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.transition1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.transition1_1_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=64, bias=True, padding_mode="zeros")
        self.transition1_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition1_1_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.transition1_1_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=64, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.transition1_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition1_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_branches_0_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage2_0_branches_0_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_0_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_branches_0_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_branches_0_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_branches_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_0_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_branches_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage2_0_branches_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_branches_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_branches_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_1_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_branches_1_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_branches_1_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_fuse_layers_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage2_0_fuse_layers_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_fuse_layers_0_1_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage2_0_relu_cbrs_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_relu_cbrs_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_relu_cbrs_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_fuse_layers_1_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage2_0_fuse_layers_1_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_fuse_layers_1_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage2_0_fuse_layers_1_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_fuse_layers_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_relu_cbrs_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage2_0_relu_cbrs_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage2_0_relu_cbrs_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.transition2_2_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.transition2_2_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition2_2_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.transition2_2_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.transition2_2_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.transition2_2_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_0_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_0_branches_0_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_0_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_0_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_0_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_0_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_0_branches_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_1_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_1_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_1_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_2_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=56, bias=True, padding_mode="zeros")
        self.stage3_0_branches_2_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_2_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_branches_2_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_2_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_2_1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_branches_2_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_branches_2_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_0_fuse_layers_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_0_1_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_0_fuse_layers_0_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_0_fuse_layers_0_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_0_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=4.0, mode="nearest", align_corners=None)
        self.stage3_0_relu_cbrs_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_relu_cbrs_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_relu_cbrs_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_1_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_1_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_1_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_1_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_1_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_0_fuse_layers_1_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_1_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_0_relu_cbrs_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_relu_cbrs_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_relu_cbrs_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_2_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_2_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_2_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_2_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_2_0_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_0_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_2_0_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_2_0_1_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_2_1_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_fuse_layers_2_1_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_0_fuse_layers_2_1_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_fuse_layers_2_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_relu_cbrs_2_0_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_0_relu_cbrs_2_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_0_relu_cbrs_2_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_0_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_1_branches_0_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_0_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_0_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_0_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_0_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_1_branches_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_1_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_1_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_1_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_2_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=56, bias=True, padding_mode="zeros")
        self.stage3_1_branches_2_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_2_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_branches_2_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_2_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_2_1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_branches_2_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_branches_2_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_1_fuse_layers_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_0_1_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_1_fuse_layers_0_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_1_fuse_layers_0_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_0_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=4.0, mode="nearest", align_corners=None)
        self.stage3_1_relu_cbrs_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_relu_cbrs_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_relu_cbrs_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_1_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_1_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_1_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_1_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_1_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_1_fuse_layers_1_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_1_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_1_relu_cbrs_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_relu_cbrs_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_relu_cbrs_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_2_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_2_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_2_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_2_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_2_0_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_0_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_2_0_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_2_0_1_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_2_1_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_fuse_layers_2_1_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_1_fuse_layers_2_1_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_fuse_layers_2_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_relu_cbrs_2_0_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_1_relu_cbrs_2_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_1_relu_cbrs_2_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_0_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_2_branches_0_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_0_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_0_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_0_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_0_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_2_branches_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_1_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_1_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_1_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_2_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=56, bias=True, padding_mode="zeros")
        self.stage3_2_branches_2_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_2_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_branches_2_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_2_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_2_1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_branches_2_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_branches_2_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_2_fuse_layers_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_0_1_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_2_fuse_layers_0_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_2_fuse_layers_0_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_0_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=4.0, mode="nearest", align_corners=None)
        self.stage3_2_relu_cbrs_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_relu_cbrs_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_relu_cbrs_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_1_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_1_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_1_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_1_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_1_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_2_fuse_layers_1_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_1_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_2_relu_cbrs_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_relu_cbrs_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_relu_cbrs_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_2_0_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_0_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_2_0_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_2_0_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_2_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_2_0_1_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_0_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_2_0_1_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_2_0_1_0_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_2_1_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_1_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_fuse_layers_2_1_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_2_fuse_layers_2_1_0_0_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_fuse_layers_2_1_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_relu_cbrs_2_0_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_2_relu_cbrs_2_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_2_relu_cbrs_2_0_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_0_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.stage3_3_branches_0_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_0_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_0_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_0_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_0_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=32, bias=True, padding_mode="zeros")
        self.stage3_3_branches_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_1_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=32, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_1_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=32, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_1_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_2_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=56, bias=True, padding_mode="zeros")
        self.stage3_3_branches_2_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_2_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_branches_2_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_2_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_2_1_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=56, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_branches_2_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=56, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_branches_2_1_2 = torch.nn.modules.activation.ReLU(False)
        self.stage3_3_fuse_layers_0_1_0 = torch.nn.modules.conv.Conv2d(in_channels=32, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_3_fuse_layers_0_1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_fuse_layers_0_1_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.stage3_3_fuse_layers_0_2_0 = torch.nn.modules.conv.Conv2d(in_channels=56, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.stage3_3_fuse_layers_0_2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_fuse_layers_0_2_2 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=4.0, mode="nearest", align_corners=None)
        self.stage3_3_relu_cbrs_0_0_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.stage3_3_relu_cbrs_0_0_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.stage3_3_relu_cbrs_0_0_2 = torch.nn.modules.activation.ReLU(False)
        self.final_layers_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=46, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.deconv_layers_0_0_0 = torch.nn.modules.upsampling.Upsample(size=None, scale_factor=2.0, mode="nearest", align_corners=None)
        self.deconv_layers_0_0_1 = torch.nn.modules.conv.Conv2d(in_channels=62, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=False, padding_mode="zeros")
        self.deconv_layers_0_0_2 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.deconv_layers_0_0_3 = torch.nn.modules.activation.ReLU(False)
        self.deconv_layers_0_1_0_conv1_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), dilation=(1, 1), groups=16, bias=True, padding_mode="zeros")
        self.deconv_layers_0_1_0_conv1_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.deconv_layers_0_1_0_conv1_2 = torch.nn.modules.activation.ReLU(False)
        self.deconv_layers_0_1_0_conv1_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.deconv_layers_0_1_0_conv1_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.deconv_layers_0_1_0_relu1 = torch.nn.modules.activation.ReLU(False)
        self.deconv_layers_0_1_0_conv2_0 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(3, 3), stride=(1, 1), padding=(3, 3), dilation=(3, 3), groups=16, bias=True, padding_mode="zeros")
        self.deconv_layers_0_1_0_conv2_1 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.deconv_layers_0_1_0_conv2_2 = torch.nn.modules.activation.ReLU(False)
        self.deconv_layers_0_1_0_conv2_3 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=16, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")
        self.deconv_layers_0_1_0_conv2_4 = torch.nn.modules.batchnorm.BatchNorm2d(num_features=16, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        self.final_layers_1 = torch.nn.modules.conv.Conv2d(in_channels=16, out_channels=23, kernel_size=(1, 1), stride=(1, 1), padding=(0, 0), dilation=(1, 1), groups=1, bias=True, padding_mode="zeros")

    def forward(self, input_1):
        conv1 = self.conv1(input_1)
        bn1 = self.bn1(conv1)
        relu1 = self.relu1(bn1)
        downsample_conv1_0 = self.downsample_conv1_0(relu1)
        downsample_conv1_1 = self.downsample_conv1_1(downsample_conv1_0)
        downsample_conv1_2 = self.downsample_conv1_2(downsample_conv1_1)
        downsample_conv1_3 = self.downsample_conv1_3(downsample_conv1_2)
        downsample_conv1_4 = self.downsample_conv1_4(downsample_conv1_3)
        downsample_relu1 = self.downsample_relu1(downsample_conv1_4)
        downsample_conv2_0 = self.downsample_conv2_0(downsample_relu1)
        downsample_conv2_1 = self.downsample_conv2_1(downsample_conv2_0)
        downsample_conv2_2 = self.downsample_conv2_2(downsample_conv2_1)
        downsample_conv2_3 = self.downsample_conv2_3(downsample_conv2_2)
        downsample_conv2_4 = self.downsample_conv2_4(downsample_conv2_3)
        downsample_downsample_res_conv_0 = self.downsample_downsample_res_conv_0(relu1)
        downsample_downsample_res_conv_1 = self.downsample_downsample_res_conv_1(downsample_downsample_res_conv_0)
        downsample_downsample_res_conv_2 = self.downsample_downsample_res_conv_2(downsample_downsample_res_conv_1)
        downsample_downsample_res_conv_3 = self.downsample_downsample_res_conv_3(downsample_downsample_res_conv_2)
        downsample_downsample_res_conv_4 = self.downsample_downsample_res_conv_4(downsample_downsample_res_conv_3)
        add_1 = torch.add(downsample_conv2_4, downsample_downsample_res_conv_4)
        downsample_relu2 = self.downsample_relu2(add_1)
        layer1_0_conv1_0 = self.layer1_0_conv1_0(downsample_relu2)
        layer1_0_conv1_1 = self.layer1_0_conv1_1(layer1_0_conv1_0)
        layer1_0_conv1_2 = self.layer1_0_conv1_2(layer1_0_conv1_1)
        layer1_0_conv1_3 = self.layer1_0_conv1_3(layer1_0_conv1_2)
        layer1_0_conv1_4 = self.layer1_0_conv1_4(layer1_0_conv1_3)
        add_2 = torch.add(layer1_0_conv1_4, downsample_relu2)
        layer1_1_conv1_0 = self.layer1_1_conv1_0(add_2)
        layer1_1_conv1_1 = self.layer1_1_conv1_1(layer1_1_conv1_0)
        layer1_1_conv1_2 = self.layer1_1_conv1_2(layer1_1_conv1_1)
        layer1_1_conv1_3 = self.layer1_1_conv1_3(layer1_1_conv1_2)
        layer1_1_conv1_4 = self.layer1_1_conv1_4(layer1_1_conv1_3)
        add_3 = torch.add(layer1_1_conv1_4, add_2)
        cut1_0 = self.cut1_0(add_3)
        cut1_1 = self.cut1_1(cut1_0)
        cut1_2 = self.cut1_2(cut1_1)
        transition1_0_0_0 = self.transition1_0_0_0(cut1_2)
        transition1_0_0_1 = self.transition1_0_0_1(transition1_0_0_0)
        transition1_0_0_2 = self.transition1_0_0_2(transition1_0_0_1)
        transition1_0_0_3 = self.transition1_0_0_3(transition1_0_0_2)
        transition1_0_1 = self.transition1_0_1(transition1_0_0_3)
        transition1_0_2 = self.transition1_0_2(transition1_0_1)
        transition1_1_0_0_0 = self.transition1_1_0_0_0(cut1_2)
        transition1_1_0_0_1 = self.transition1_1_0_0_1(transition1_1_0_0_0)
        transition1_1_0_0_2 = self.transition1_1_0_0_2(transition1_1_0_0_1)
        transition1_1_0_0_3 = self.transition1_1_0_0_3(transition1_1_0_0_2)
        transition1_1_0_1 = self.transition1_1_0_1(transition1_1_0_0_3)
        transition1_1_0_2 = self.transition1_1_0_2(transition1_1_0_1)
        stage2_0_branches_0_0_conv1_0 = self.stage2_0_branches_0_0_conv1_0(transition1_0_2)
        stage2_0_branches_0_0_conv1_1 = self.stage2_0_branches_0_0_conv1_1(stage2_0_branches_0_0_conv1_0)
        stage2_0_branches_0_0_conv1_2 = self.stage2_0_branches_0_0_conv1_2(stage2_0_branches_0_0_conv1_1)
        stage2_0_branches_0_0_conv1_3 = self.stage2_0_branches_0_0_conv1_3(stage2_0_branches_0_0_conv1_2)
        stage2_0_branches_0_0_conv1_4 = self.stage2_0_branches_0_0_conv1_4(stage2_0_branches_0_0_conv1_3)
        add_4 = torch.add(stage2_0_branches_0_0_conv1_4, transition1_0_2)
        stage2_0_branches_0_1_0 = self.stage2_0_branches_0_1_0(add_4)
        stage2_0_branches_0_1_1 = self.stage2_0_branches_0_1_1(stage2_0_branches_0_1_0)
        stage2_0_branches_0_1_2 = self.stage2_0_branches_0_1_2(stage2_0_branches_0_1_1)
        stage2_0_branches_1_0_conv1_0 = self.stage2_0_branches_1_0_conv1_0(transition1_1_0_2)
        stage2_0_branches_1_0_conv1_1 = self.stage2_0_branches_1_0_conv1_1(stage2_0_branches_1_0_conv1_0)
        stage2_0_branches_1_0_conv1_2 = self.stage2_0_branches_1_0_conv1_2(stage2_0_branches_1_0_conv1_1)
        stage2_0_branches_1_0_conv1_3 = self.stage2_0_branches_1_0_conv1_3(stage2_0_branches_1_0_conv1_2)
        stage2_0_branches_1_0_conv1_4 = self.stage2_0_branches_1_0_conv1_4(stage2_0_branches_1_0_conv1_3)
        add_5 = torch.add(stage2_0_branches_1_0_conv1_4, transition1_1_0_2)
        stage2_0_branches_1_1_0 = self.stage2_0_branches_1_1_0(add_5)
        stage2_0_branches_1_1_1 = self.stage2_0_branches_1_1_1(stage2_0_branches_1_1_0)
        stage2_0_branches_1_1_2 = self.stage2_0_branches_1_1_2(stage2_0_branches_1_1_1)
        stage2_0_fuse_layers_0_1_0 = self.stage2_0_fuse_layers_0_1_0(stage2_0_branches_1_1_2)
        stage2_0_fuse_layers_0_1_1 = self.stage2_0_fuse_layers_0_1_1(stage2_0_fuse_layers_0_1_0)
        stage2_0_fuse_layers_0_1_2 = self.stage2_0_fuse_layers_0_1_2(stage2_0_fuse_layers_0_1_1)
        add_6 = torch.add(stage2_0_branches_0_1_2, stage2_0_fuse_layers_0_1_2)
        stage2_0_relu_cbrs_0_0_0 = self.stage2_0_relu_cbrs_0_0_0(add_6)
        stage2_0_relu_cbrs_0_0_1 = self.stage2_0_relu_cbrs_0_0_1(stage2_0_relu_cbrs_0_0_0)
        stage2_0_relu_cbrs_0_0_2 = self.stage2_0_relu_cbrs_0_0_2(stage2_0_relu_cbrs_0_0_1)
        stage2_0_fuse_layers_1_0_0_0_0 = self.stage2_0_fuse_layers_1_0_0_0_0(stage2_0_branches_0_1_2)
        stage2_0_fuse_layers_1_0_0_0_1 = self.stage2_0_fuse_layers_1_0_0_0_1(stage2_0_fuse_layers_1_0_0_0_0)
        stage2_0_fuse_layers_1_0_0_0_2 = self.stage2_0_fuse_layers_1_0_0_0_2(stage2_0_fuse_layers_1_0_0_0_1)
        stage2_0_fuse_layers_1_0_0_0_3 = self.stage2_0_fuse_layers_1_0_0_0_3(stage2_0_fuse_layers_1_0_0_0_2)
        stage2_0_fuse_layers_1_0_0_1 = self.stage2_0_fuse_layers_1_0_0_1(stage2_0_fuse_layers_1_0_0_0_3)
        add_7 = torch.add(stage2_0_fuse_layers_1_0_0_1, stage2_0_branches_1_1_2)
        stage2_0_relu_cbrs_1_0_0 = self.stage2_0_relu_cbrs_1_0_0(add_7)
        stage2_0_relu_cbrs_1_0_1 = self.stage2_0_relu_cbrs_1_0_1(stage2_0_relu_cbrs_1_0_0)
        stage2_0_relu_cbrs_1_0_2 = self.stage2_0_relu_cbrs_1_0_2(stage2_0_relu_cbrs_1_0_1)
        transition2_2_0_0_0 = self.transition2_2_0_0_0(stage2_0_relu_cbrs_1_0_2)
        transition2_2_0_0_1 = self.transition2_2_0_0_1(transition2_2_0_0_0)
        transition2_2_0_0_2 = self.transition2_2_0_0_2(transition2_2_0_0_1)
        transition2_2_0_0_3 = self.transition2_2_0_0_3(transition2_2_0_0_2)
        transition2_2_0_1 = self.transition2_2_0_1(transition2_2_0_0_3)
        transition2_2_0_2 = self.transition2_2_0_2(transition2_2_0_1)
        stage3_0_branches_0_0_conv1_0 = self.stage3_0_branches_0_0_conv1_0(stage2_0_relu_cbrs_0_0_2)
        stage3_0_branches_0_0_conv1_1 = self.stage3_0_branches_0_0_conv1_1(stage3_0_branches_0_0_conv1_0)
        stage3_0_branches_0_0_conv1_2 = self.stage3_0_branches_0_0_conv1_2(stage3_0_branches_0_0_conv1_1)
        stage3_0_branches_0_0_conv1_3 = self.stage3_0_branches_0_0_conv1_3(stage3_0_branches_0_0_conv1_2)
        stage3_0_branches_0_0_conv1_4 = self.stage3_0_branches_0_0_conv1_4(stage3_0_branches_0_0_conv1_3)
        add_8 = torch.add(stage3_0_branches_0_0_conv1_4, stage2_0_relu_cbrs_0_0_2)
        stage3_0_branches_0_1_0 = self.stage3_0_branches_0_1_0(add_8)
        stage3_0_branches_0_1_1 = self.stage3_0_branches_0_1_1(stage3_0_branches_0_1_0)
        stage3_0_branches_0_1_2 = self.stage3_0_branches_0_1_2(stage3_0_branches_0_1_1)
        stage3_0_branches_1_0_conv1_0 = self.stage3_0_branches_1_0_conv1_0(stage2_0_relu_cbrs_1_0_2)
        stage3_0_branches_1_0_conv1_1 = self.stage3_0_branches_1_0_conv1_1(stage3_0_branches_1_0_conv1_0)
        stage3_0_branches_1_0_conv1_2 = self.stage3_0_branches_1_0_conv1_2(stage3_0_branches_1_0_conv1_1)
        stage3_0_branches_1_0_conv1_3 = self.stage3_0_branches_1_0_conv1_3(stage3_0_branches_1_0_conv1_2)
        stage3_0_branches_1_0_conv1_4 = self.stage3_0_branches_1_0_conv1_4(stage3_0_branches_1_0_conv1_3)
        add_9 = torch.add(stage3_0_branches_1_0_conv1_4, stage2_0_relu_cbrs_1_0_2)
        stage3_0_branches_1_1_0 = self.stage3_0_branches_1_1_0(add_9)
        stage3_0_branches_1_1_1 = self.stage3_0_branches_1_1_1(stage3_0_branches_1_1_0)
        stage3_0_branches_1_1_2 = self.stage3_0_branches_1_1_2(stage3_0_branches_1_1_1)
        stage3_0_branches_2_0_conv1_0 = self.stage3_0_branches_2_0_conv1_0(transition2_2_0_2)
        stage3_0_branches_2_0_conv1_1 = self.stage3_0_branches_2_0_conv1_1(stage3_0_branches_2_0_conv1_0)
        stage3_0_branches_2_0_conv1_2 = self.stage3_0_branches_2_0_conv1_2(stage3_0_branches_2_0_conv1_1)
        stage3_0_branches_2_0_conv1_3 = self.stage3_0_branches_2_0_conv1_3(stage3_0_branches_2_0_conv1_2)
        stage3_0_branches_2_0_conv1_4 = self.stage3_0_branches_2_0_conv1_4(stage3_0_branches_2_0_conv1_3)
        add_10 = torch.add(stage3_0_branches_2_0_conv1_4, transition2_2_0_2)
        stage3_0_branches_2_1_0 = self.stage3_0_branches_2_1_0(add_10)
        stage3_0_branches_2_1_1 = self.stage3_0_branches_2_1_1(stage3_0_branches_2_1_0)
        stage3_0_branches_2_1_2 = self.stage3_0_branches_2_1_2(stage3_0_branches_2_1_1)
        stage3_0_fuse_layers_0_1_0 = self.stage3_0_fuse_layers_0_1_0(stage3_0_branches_1_1_2)
        stage3_0_fuse_layers_0_1_1 = self.stage3_0_fuse_layers_0_1_1(stage3_0_fuse_layers_0_1_0)
        stage3_0_fuse_layers_0_1_2 = self.stage3_0_fuse_layers_0_1_2(stage3_0_fuse_layers_0_1_1)
        add_11 = torch.add(stage3_0_branches_0_1_2, stage3_0_fuse_layers_0_1_2)
        stage3_0_fuse_layers_0_2_0 = self.stage3_0_fuse_layers_0_2_0(stage3_0_branches_2_1_2)
        stage3_0_fuse_layers_0_2_1 = self.stage3_0_fuse_layers_0_2_1(stage3_0_fuse_layers_0_2_0)
        stage3_0_fuse_layers_0_2_2 = self.stage3_0_fuse_layers_0_2_2(stage3_0_fuse_layers_0_2_1)
        add_12 = torch.add(add_11, stage3_0_fuse_layers_0_2_2)
        stage3_0_relu_cbrs_0_0_0 = self.stage3_0_relu_cbrs_0_0_0(add_12)
        stage3_0_relu_cbrs_0_0_1 = self.stage3_0_relu_cbrs_0_0_1(stage3_0_relu_cbrs_0_0_0)
        stage3_0_relu_cbrs_0_0_2 = self.stage3_0_relu_cbrs_0_0_2(stage3_0_relu_cbrs_0_0_1)
        stage3_0_fuse_layers_1_0_0_0_0 = self.stage3_0_fuse_layers_1_0_0_0_0(stage3_0_branches_0_1_2)
        stage3_0_fuse_layers_1_0_0_0_1 = self.stage3_0_fuse_layers_1_0_0_0_1(stage3_0_fuse_layers_1_0_0_0_0)
        stage3_0_fuse_layers_1_0_0_0_2 = self.stage3_0_fuse_layers_1_0_0_0_2(stage3_0_fuse_layers_1_0_0_0_1)
        stage3_0_fuse_layers_1_0_0_0_3 = self.stage3_0_fuse_layers_1_0_0_0_3(stage3_0_fuse_layers_1_0_0_0_2)
        stage3_0_fuse_layers_1_0_0_1 = self.stage3_0_fuse_layers_1_0_0_1(stage3_0_fuse_layers_1_0_0_0_3)
        add_13 = torch.add(stage3_0_fuse_layers_1_0_0_1, stage3_0_branches_1_1_2)
        stage3_0_fuse_layers_1_2_0 = self.stage3_0_fuse_layers_1_2_0(stage3_0_branches_2_1_2)
        stage3_0_fuse_layers_1_2_1 = self.stage3_0_fuse_layers_1_2_1(stage3_0_fuse_layers_1_2_0)
        stage3_0_fuse_layers_1_2_2 = self.stage3_0_fuse_layers_1_2_2(stage3_0_fuse_layers_1_2_1)
        add_14 = torch.add(add_13, stage3_0_fuse_layers_1_2_2)
        stage3_0_relu_cbrs_1_0_0 = self.stage3_0_relu_cbrs_1_0_0(add_14)
        stage3_0_relu_cbrs_1_0_1 = self.stage3_0_relu_cbrs_1_0_1(stage3_0_relu_cbrs_1_0_0)
        stage3_0_relu_cbrs_1_0_2 = self.stage3_0_relu_cbrs_1_0_2(stage3_0_relu_cbrs_1_0_1)
        stage3_0_fuse_layers_2_0_0_0_0 = self.stage3_0_fuse_layers_2_0_0_0_0(stage3_0_branches_0_1_2)
        stage3_0_fuse_layers_2_0_0_0_1 = self.stage3_0_fuse_layers_2_0_0_0_1(stage3_0_fuse_layers_2_0_0_0_0)
        stage3_0_fuse_layers_2_0_0_0_2 = self.stage3_0_fuse_layers_2_0_0_0_2(stage3_0_fuse_layers_2_0_0_0_1)
        stage3_0_fuse_layers_2_0_0_0_3 = self.stage3_0_fuse_layers_2_0_0_0_3(stage3_0_fuse_layers_2_0_0_0_2)
        stage3_0_fuse_layers_2_0_0_1 = self.stage3_0_fuse_layers_2_0_0_1(stage3_0_fuse_layers_2_0_0_0_3)
        stage3_0_fuse_layers_2_0_0_2 = self.stage3_0_fuse_layers_2_0_0_2(stage3_0_fuse_layers_2_0_0_1)
        stage3_0_fuse_layers_2_0_1_0_0 = self.stage3_0_fuse_layers_2_0_1_0_0(stage3_0_fuse_layers_2_0_0_2)
        stage3_0_fuse_layers_2_0_1_0_1 = self.stage3_0_fuse_layers_2_0_1_0_1(stage3_0_fuse_layers_2_0_1_0_0)
        stage3_0_fuse_layers_2_0_1_0_2 = self.stage3_0_fuse_layers_2_0_1_0_2(stage3_0_fuse_layers_2_0_1_0_1)
        stage3_0_fuse_layers_2_0_1_0_3 = self.stage3_0_fuse_layers_2_0_1_0_3(stage3_0_fuse_layers_2_0_1_0_2)
        stage3_0_fuse_layers_2_0_1_1 = self.stage3_0_fuse_layers_2_0_1_1(stage3_0_fuse_layers_2_0_1_0_3)
        stage3_0_fuse_layers_2_1_0_0_0 = self.stage3_0_fuse_layers_2_1_0_0_0(stage3_0_branches_1_1_2)
        stage3_0_fuse_layers_2_1_0_0_1 = self.stage3_0_fuse_layers_2_1_0_0_1(stage3_0_fuse_layers_2_1_0_0_0)
        stage3_0_fuse_layers_2_1_0_0_2 = self.stage3_0_fuse_layers_2_1_0_0_2(stage3_0_fuse_layers_2_1_0_0_1)
        stage3_0_fuse_layers_2_1_0_0_3 = self.stage3_0_fuse_layers_2_1_0_0_3(stage3_0_fuse_layers_2_1_0_0_2)
        stage3_0_fuse_layers_2_1_0_1 = self.stage3_0_fuse_layers_2_1_0_1(stage3_0_fuse_layers_2_1_0_0_3)
        add_15 = torch.add(stage3_0_fuse_layers_2_0_1_1, stage3_0_fuse_layers_2_1_0_1)
        add_16 = torch.add(add_15, stage3_0_branches_2_1_2)
        stage3_0_relu_cbrs_2_0_0 = self.stage3_0_relu_cbrs_2_0_0(add_16)
        stage3_0_relu_cbrs_2_0_1 = self.stage3_0_relu_cbrs_2_0_1(stage3_0_relu_cbrs_2_0_0)
        stage3_0_relu_cbrs_2_0_2 = self.stage3_0_relu_cbrs_2_0_2(stage3_0_relu_cbrs_2_0_1)
        stage3_1_branches_0_0_conv1_0 = self.stage3_1_branches_0_0_conv1_0(stage3_0_relu_cbrs_0_0_2)
        stage3_1_branches_0_0_conv1_1 = self.stage3_1_branches_0_0_conv1_1(stage3_1_branches_0_0_conv1_0)
        stage3_1_branches_0_0_conv1_2 = self.stage3_1_branches_0_0_conv1_2(stage3_1_branches_0_0_conv1_1)
        stage3_1_branches_0_0_conv1_3 = self.stage3_1_branches_0_0_conv1_3(stage3_1_branches_0_0_conv1_2)
        stage3_1_branches_0_0_conv1_4 = self.stage3_1_branches_0_0_conv1_4(stage3_1_branches_0_0_conv1_3)
        add_17 = torch.add(stage3_1_branches_0_0_conv1_4, stage3_0_relu_cbrs_0_0_2)
        stage3_1_branches_0_1_0 = self.stage3_1_branches_0_1_0(add_17)
        stage3_1_branches_0_1_1 = self.stage3_1_branches_0_1_1(stage3_1_branches_0_1_0)
        stage3_1_branches_0_1_2 = self.stage3_1_branches_0_1_2(stage3_1_branches_0_1_1)
        stage3_1_branches_1_0_conv1_0 = self.stage3_1_branches_1_0_conv1_0(stage3_0_relu_cbrs_1_0_2)
        stage3_1_branches_1_0_conv1_1 = self.stage3_1_branches_1_0_conv1_1(stage3_1_branches_1_0_conv1_0)
        stage3_1_branches_1_0_conv1_2 = self.stage3_1_branches_1_0_conv1_2(stage3_1_branches_1_0_conv1_1)
        stage3_1_branches_1_0_conv1_3 = self.stage3_1_branches_1_0_conv1_3(stage3_1_branches_1_0_conv1_2)
        stage3_1_branches_1_0_conv1_4 = self.stage3_1_branches_1_0_conv1_4(stage3_1_branches_1_0_conv1_3)
        add_18 = torch.add(stage3_1_branches_1_0_conv1_4, stage3_0_relu_cbrs_1_0_2)
        stage3_1_branches_1_1_0 = self.stage3_1_branches_1_1_0(add_18)
        stage3_1_branches_1_1_1 = self.stage3_1_branches_1_1_1(stage3_1_branches_1_1_0)
        stage3_1_branches_1_1_2 = self.stage3_1_branches_1_1_2(stage3_1_branches_1_1_1)
        stage3_1_branches_2_0_conv1_0 = self.stage3_1_branches_2_0_conv1_0(stage3_0_relu_cbrs_2_0_2)
        stage3_1_branches_2_0_conv1_1 = self.stage3_1_branches_2_0_conv1_1(stage3_1_branches_2_0_conv1_0)
        stage3_1_branches_2_0_conv1_2 = self.stage3_1_branches_2_0_conv1_2(stage3_1_branches_2_0_conv1_1)
        stage3_1_branches_2_0_conv1_3 = self.stage3_1_branches_2_0_conv1_3(stage3_1_branches_2_0_conv1_2)
        stage3_1_branches_2_0_conv1_4 = self.stage3_1_branches_2_0_conv1_4(stage3_1_branches_2_0_conv1_3)
        add_19 = torch.add(stage3_1_branches_2_0_conv1_4, stage3_0_relu_cbrs_2_0_2)
        stage3_1_branches_2_1_0 = self.stage3_1_branches_2_1_0(add_19)
        stage3_1_branches_2_1_1 = self.stage3_1_branches_2_1_1(stage3_1_branches_2_1_0)
        stage3_1_branches_2_1_2 = self.stage3_1_branches_2_1_2(stage3_1_branches_2_1_1)
        stage3_1_fuse_layers_0_1_0 = self.stage3_1_fuse_layers_0_1_0(stage3_1_branches_1_1_2)
        stage3_1_fuse_layers_0_1_1 = self.stage3_1_fuse_layers_0_1_1(stage3_1_fuse_layers_0_1_0)
        stage3_1_fuse_layers_0_1_2 = self.stage3_1_fuse_layers_0_1_2(stage3_1_fuse_layers_0_1_1)
        add_20 = torch.add(stage3_1_branches_0_1_2, stage3_1_fuse_layers_0_1_2)
        stage3_1_fuse_layers_0_2_0 = self.stage3_1_fuse_layers_0_2_0(stage3_1_branches_2_1_2)
        stage3_1_fuse_layers_0_2_1 = self.stage3_1_fuse_layers_0_2_1(stage3_1_fuse_layers_0_2_0)
        stage3_1_fuse_layers_0_2_2 = self.stage3_1_fuse_layers_0_2_2(stage3_1_fuse_layers_0_2_1)
        add_21 = torch.add(add_20, stage3_1_fuse_layers_0_2_2)
        stage3_1_relu_cbrs_0_0_0 = self.stage3_1_relu_cbrs_0_0_0(add_21)
        stage3_1_relu_cbrs_0_0_1 = self.stage3_1_relu_cbrs_0_0_1(stage3_1_relu_cbrs_0_0_0)
        stage3_1_relu_cbrs_0_0_2 = self.stage3_1_relu_cbrs_0_0_2(stage3_1_relu_cbrs_0_0_1)
        stage3_1_fuse_layers_1_0_0_0_0 = self.stage3_1_fuse_layers_1_0_0_0_0(stage3_1_branches_0_1_2)
        stage3_1_fuse_layers_1_0_0_0_1 = self.stage3_1_fuse_layers_1_0_0_0_1(stage3_1_fuse_layers_1_0_0_0_0)
        stage3_1_fuse_layers_1_0_0_0_2 = self.stage3_1_fuse_layers_1_0_0_0_2(stage3_1_fuse_layers_1_0_0_0_1)
        stage3_1_fuse_layers_1_0_0_0_3 = self.stage3_1_fuse_layers_1_0_0_0_3(stage3_1_fuse_layers_1_0_0_0_2)
        stage3_1_fuse_layers_1_0_0_1 = self.stage3_1_fuse_layers_1_0_0_1(stage3_1_fuse_layers_1_0_0_0_3)
        add_22 = torch.add(stage3_1_fuse_layers_1_0_0_1, stage3_1_branches_1_1_2)
        stage3_1_fuse_layers_1_2_0 = self.stage3_1_fuse_layers_1_2_0(stage3_1_branches_2_1_2)
        stage3_1_fuse_layers_1_2_1 = self.stage3_1_fuse_layers_1_2_1(stage3_1_fuse_layers_1_2_0)
        stage3_1_fuse_layers_1_2_2 = self.stage3_1_fuse_layers_1_2_2(stage3_1_fuse_layers_1_2_1)
        add_23 = torch.add(add_22, stage3_1_fuse_layers_1_2_2)
        stage3_1_relu_cbrs_1_0_0 = self.stage3_1_relu_cbrs_1_0_0(add_23)
        stage3_1_relu_cbrs_1_0_1 = self.stage3_1_relu_cbrs_1_0_1(stage3_1_relu_cbrs_1_0_0)
        stage3_1_relu_cbrs_1_0_2 = self.stage3_1_relu_cbrs_1_0_2(stage3_1_relu_cbrs_1_0_1)
        stage3_1_fuse_layers_2_0_0_0_0 = self.stage3_1_fuse_layers_2_0_0_0_0(stage3_1_branches_0_1_2)
        stage3_1_fuse_layers_2_0_0_0_1 = self.stage3_1_fuse_layers_2_0_0_0_1(stage3_1_fuse_layers_2_0_0_0_0)
        stage3_1_fuse_layers_2_0_0_0_2 = self.stage3_1_fuse_layers_2_0_0_0_2(stage3_1_fuse_layers_2_0_0_0_1)
        stage3_1_fuse_layers_2_0_0_0_3 = self.stage3_1_fuse_layers_2_0_0_0_3(stage3_1_fuse_layers_2_0_0_0_2)
        stage3_1_fuse_layers_2_0_0_1 = self.stage3_1_fuse_layers_2_0_0_1(stage3_1_fuse_layers_2_0_0_0_3)
        stage3_1_fuse_layers_2_0_0_2 = self.stage3_1_fuse_layers_2_0_0_2(stage3_1_fuse_layers_2_0_0_1)
        stage3_1_fuse_layers_2_0_1_0_0 = self.stage3_1_fuse_layers_2_0_1_0_0(stage3_1_fuse_layers_2_0_0_2)
        stage3_1_fuse_layers_2_0_1_0_1 = self.stage3_1_fuse_layers_2_0_1_0_1(stage3_1_fuse_layers_2_0_1_0_0)
        stage3_1_fuse_layers_2_0_1_0_2 = self.stage3_1_fuse_layers_2_0_1_0_2(stage3_1_fuse_layers_2_0_1_0_1)
        stage3_1_fuse_layers_2_0_1_0_3 = self.stage3_1_fuse_layers_2_0_1_0_3(stage3_1_fuse_layers_2_0_1_0_2)
        stage3_1_fuse_layers_2_0_1_1 = self.stage3_1_fuse_layers_2_0_1_1(stage3_1_fuse_layers_2_0_1_0_3)
        stage3_1_fuse_layers_2_1_0_0_0 = self.stage3_1_fuse_layers_2_1_0_0_0(stage3_1_branches_1_1_2)
        stage3_1_fuse_layers_2_1_0_0_1 = self.stage3_1_fuse_layers_2_1_0_0_1(stage3_1_fuse_layers_2_1_0_0_0)
        stage3_1_fuse_layers_2_1_0_0_2 = self.stage3_1_fuse_layers_2_1_0_0_2(stage3_1_fuse_layers_2_1_0_0_1)
        stage3_1_fuse_layers_2_1_0_0_3 = self.stage3_1_fuse_layers_2_1_0_0_3(stage3_1_fuse_layers_2_1_0_0_2)
        stage3_1_fuse_layers_2_1_0_1 = self.stage3_1_fuse_layers_2_1_0_1(stage3_1_fuse_layers_2_1_0_0_3)
        add_24 = torch.add(stage3_1_fuse_layers_2_0_1_1, stage3_1_fuse_layers_2_1_0_1)
        add_25 = torch.add(add_24, stage3_1_branches_2_1_2)
        stage3_1_relu_cbrs_2_0_0 = self.stage3_1_relu_cbrs_2_0_0(add_25)
        stage3_1_relu_cbrs_2_0_1 = self.stage3_1_relu_cbrs_2_0_1(stage3_1_relu_cbrs_2_0_0)
        stage3_1_relu_cbrs_2_0_2 = self.stage3_1_relu_cbrs_2_0_2(stage3_1_relu_cbrs_2_0_1)
        stage3_2_branches_0_0_conv1_0 = self.stage3_2_branches_0_0_conv1_0(stage3_1_relu_cbrs_0_0_2)
        stage3_2_branches_0_0_conv1_1 = self.stage3_2_branches_0_0_conv1_1(stage3_2_branches_0_0_conv1_0)
        stage3_2_branches_0_0_conv1_2 = self.stage3_2_branches_0_0_conv1_2(stage3_2_branches_0_0_conv1_1)
        stage3_2_branches_0_0_conv1_3 = self.stage3_2_branches_0_0_conv1_3(stage3_2_branches_0_0_conv1_2)
        stage3_2_branches_0_0_conv1_4 = self.stage3_2_branches_0_0_conv1_4(stage3_2_branches_0_0_conv1_3)
        add_26 = torch.add(stage3_2_branches_0_0_conv1_4, stage3_1_relu_cbrs_0_0_2)
        stage3_2_branches_0_1_0 = self.stage3_2_branches_0_1_0(add_26)
        stage3_2_branches_0_1_1 = self.stage3_2_branches_0_1_1(stage3_2_branches_0_1_0)
        stage3_2_branches_0_1_2 = self.stage3_2_branches_0_1_2(stage3_2_branches_0_1_1)
        stage3_2_branches_1_0_conv1_0 = self.stage3_2_branches_1_0_conv1_0(stage3_1_relu_cbrs_1_0_2)
        stage3_2_branches_1_0_conv1_1 = self.stage3_2_branches_1_0_conv1_1(stage3_2_branches_1_0_conv1_0)
        stage3_2_branches_1_0_conv1_2 = self.stage3_2_branches_1_0_conv1_2(stage3_2_branches_1_0_conv1_1)
        stage3_2_branches_1_0_conv1_3 = self.stage3_2_branches_1_0_conv1_3(stage3_2_branches_1_0_conv1_2)
        stage3_2_branches_1_0_conv1_4 = self.stage3_2_branches_1_0_conv1_4(stage3_2_branches_1_0_conv1_3)
        add_27 = torch.add(stage3_2_branches_1_0_conv1_4, stage3_1_relu_cbrs_1_0_2)
        stage3_2_branches_1_1_0 = self.stage3_2_branches_1_1_0(add_27)
        stage3_2_branches_1_1_1 = self.stage3_2_branches_1_1_1(stage3_2_branches_1_1_0)
        stage3_2_branches_1_1_2 = self.stage3_2_branches_1_1_2(stage3_2_branches_1_1_1)
        stage3_2_branches_2_0_conv1_0 = self.stage3_2_branches_2_0_conv1_0(stage3_1_relu_cbrs_2_0_2)
        stage3_2_branches_2_0_conv1_1 = self.stage3_2_branches_2_0_conv1_1(stage3_2_branches_2_0_conv1_0)
        stage3_2_branches_2_0_conv1_2 = self.stage3_2_branches_2_0_conv1_2(stage3_2_branches_2_0_conv1_1)
        stage3_2_branches_2_0_conv1_3 = self.stage3_2_branches_2_0_conv1_3(stage3_2_branches_2_0_conv1_2)
        stage3_2_branches_2_0_conv1_4 = self.stage3_2_branches_2_0_conv1_4(stage3_2_branches_2_0_conv1_3)
        add_28 = torch.add(stage3_2_branches_2_0_conv1_4, stage3_1_relu_cbrs_2_0_2)
        stage3_2_branches_2_1_0 = self.stage3_2_branches_2_1_0(add_28)
        stage3_2_branches_2_1_1 = self.stage3_2_branches_2_1_1(stage3_2_branches_2_1_0)
        stage3_2_branches_2_1_2 = self.stage3_2_branches_2_1_2(stage3_2_branches_2_1_1)
        stage3_2_fuse_layers_0_1_0 = self.stage3_2_fuse_layers_0_1_0(stage3_2_branches_1_1_2)
        stage3_2_fuse_layers_0_1_1 = self.stage3_2_fuse_layers_0_1_1(stage3_2_fuse_layers_0_1_0)
        stage3_2_fuse_layers_0_1_2 = self.stage3_2_fuse_layers_0_1_2(stage3_2_fuse_layers_0_1_1)
        add_29 = torch.add(stage3_2_branches_0_1_2, stage3_2_fuse_layers_0_1_2)
        stage3_2_fuse_layers_0_2_0 = self.stage3_2_fuse_layers_0_2_0(stage3_2_branches_2_1_2)
        stage3_2_fuse_layers_0_2_1 = self.stage3_2_fuse_layers_0_2_1(stage3_2_fuse_layers_0_2_0)
        stage3_2_fuse_layers_0_2_2 = self.stage3_2_fuse_layers_0_2_2(stage3_2_fuse_layers_0_2_1)
        add_30 = torch.add(add_29, stage3_2_fuse_layers_0_2_2)
        stage3_2_relu_cbrs_0_0_0 = self.stage3_2_relu_cbrs_0_0_0(add_30)
        stage3_2_relu_cbrs_0_0_1 = self.stage3_2_relu_cbrs_0_0_1(stage3_2_relu_cbrs_0_0_0)
        stage3_2_relu_cbrs_0_0_2 = self.stage3_2_relu_cbrs_0_0_2(stage3_2_relu_cbrs_0_0_1)
        stage3_2_fuse_layers_1_0_0_0_0 = self.stage3_2_fuse_layers_1_0_0_0_0(stage3_2_branches_0_1_2)
        stage3_2_fuse_layers_1_0_0_0_1 = self.stage3_2_fuse_layers_1_0_0_0_1(stage3_2_fuse_layers_1_0_0_0_0)
        stage3_2_fuse_layers_1_0_0_0_2 = self.stage3_2_fuse_layers_1_0_0_0_2(stage3_2_fuse_layers_1_0_0_0_1)
        stage3_2_fuse_layers_1_0_0_0_3 = self.stage3_2_fuse_layers_1_0_0_0_3(stage3_2_fuse_layers_1_0_0_0_2)
        stage3_2_fuse_layers_1_0_0_1 = self.stage3_2_fuse_layers_1_0_0_1(stage3_2_fuse_layers_1_0_0_0_3)
        add_31 = torch.add(stage3_2_fuse_layers_1_0_0_1, stage3_2_branches_1_1_2)
        stage3_2_fuse_layers_1_2_0 = self.stage3_2_fuse_layers_1_2_0(stage3_2_branches_2_1_2)
        stage3_2_fuse_layers_1_2_1 = self.stage3_2_fuse_layers_1_2_1(stage3_2_fuse_layers_1_2_0)
        stage3_2_fuse_layers_1_2_2 = self.stage3_2_fuse_layers_1_2_2(stage3_2_fuse_layers_1_2_1)
        add_32 = torch.add(add_31, stage3_2_fuse_layers_1_2_2)
        stage3_2_relu_cbrs_1_0_0 = self.stage3_2_relu_cbrs_1_0_0(add_32)
        stage3_2_relu_cbrs_1_0_1 = self.stage3_2_relu_cbrs_1_0_1(stage3_2_relu_cbrs_1_0_0)
        stage3_2_relu_cbrs_1_0_2 = self.stage3_2_relu_cbrs_1_0_2(stage3_2_relu_cbrs_1_0_1)
        stage3_2_fuse_layers_2_0_0_0_0 = self.stage3_2_fuse_layers_2_0_0_0_0(stage3_2_branches_0_1_2)
        stage3_2_fuse_layers_2_0_0_0_1 = self.stage3_2_fuse_layers_2_0_0_0_1(stage3_2_fuse_layers_2_0_0_0_0)
        stage3_2_fuse_layers_2_0_0_0_2 = self.stage3_2_fuse_layers_2_0_0_0_2(stage3_2_fuse_layers_2_0_0_0_1)
        stage3_2_fuse_layers_2_0_0_0_3 = self.stage3_2_fuse_layers_2_0_0_0_3(stage3_2_fuse_layers_2_0_0_0_2)
        stage3_2_fuse_layers_2_0_0_1 = self.stage3_2_fuse_layers_2_0_0_1(stage3_2_fuse_layers_2_0_0_0_3)
        stage3_2_fuse_layers_2_0_0_2 = self.stage3_2_fuse_layers_2_0_0_2(stage3_2_fuse_layers_2_0_0_1)
        stage3_2_fuse_layers_2_0_1_0_0 = self.stage3_2_fuse_layers_2_0_1_0_0(stage3_2_fuse_layers_2_0_0_2)
        stage3_2_fuse_layers_2_0_1_0_1 = self.stage3_2_fuse_layers_2_0_1_0_1(stage3_2_fuse_layers_2_0_1_0_0)
        stage3_2_fuse_layers_2_0_1_0_2 = self.stage3_2_fuse_layers_2_0_1_0_2(stage3_2_fuse_layers_2_0_1_0_1)
        stage3_2_fuse_layers_2_0_1_0_3 = self.stage3_2_fuse_layers_2_0_1_0_3(stage3_2_fuse_layers_2_0_1_0_2)
        stage3_2_fuse_layers_2_0_1_1 = self.stage3_2_fuse_layers_2_0_1_1(stage3_2_fuse_layers_2_0_1_0_3)
        stage3_2_fuse_layers_2_1_0_0_0 = self.stage3_2_fuse_layers_2_1_0_0_0(stage3_2_branches_1_1_2)
        stage3_2_fuse_layers_2_1_0_0_1 = self.stage3_2_fuse_layers_2_1_0_0_1(stage3_2_fuse_layers_2_1_0_0_0)
        stage3_2_fuse_layers_2_1_0_0_2 = self.stage3_2_fuse_layers_2_1_0_0_2(stage3_2_fuse_layers_2_1_0_0_1)
        stage3_2_fuse_layers_2_1_0_0_3 = self.stage3_2_fuse_layers_2_1_0_0_3(stage3_2_fuse_layers_2_1_0_0_2)
        stage3_2_fuse_layers_2_1_0_1 = self.stage3_2_fuse_layers_2_1_0_1(stage3_2_fuse_layers_2_1_0_0_3)
        add_33 = torch.add(stage3_2_fuse_layers_2_0_1_1, stage3_2_fuse_layers_2_1_0_1)
        add_34 = torch.add(add_33, stage3_2_branches_2_1_2)
        stage3_2_relu_cbrs_2_0_0 = self.stage3_2_relu_cbrs_2_0_0(add_34)
        stage3_2_relu_cbrs_2_0_1 = self.stage3_2_relu_cbrs_2_0_1(stage3_2_relu_cbrs_2_0_0)
        stage3_2_relu_cbrs_2_0_2 = self.stage3_2_relu_cbrs_2_0_2(stage3_2_relu_cbrs_2_0_1)
        stage3_3_branches_0_0_conv1_0 = self.stage3_3_branches_0_0_conv1_0(stage3_2_relu_cbrs_0_0_2)
        stage3_3_branches_0_0_conv1_1 = self.stage3_3_branches_0_0_conv1_1(stage3_3_branches_0_0_conv1_0)
        stage3_3_branches_0_0_conv1_2 = self.stage3_3_branches_0_0_conv1_2(stage3_3_branches_0_0_conv1_1)
        stage3_3_branches_0_0_conv1_3 = self.stage3_3_branches_0_0_conv1_3(stage3_3_branches_0_0_conv1_2)
        stage3_3_branches_0_0_conv1_4 = self.stage3_3_branches_0_0_conv1_4(stage3_3_branches_0_0_conv1_3)
        add_35 = torch.add(stage3_3_branches_0_0_conv1_4, stage3_2_relu_cbrs_0_0_2)
        stage3_3_branches_0_1_0 = self.stage3_3_branches_0_1_0(add_35)
        stage3_3_branches_0_1_1 = self.stage3_3_branches_0_1_1(stage3_3_branches_0_1_0)
        stage3_3_branches_0_1_2 = self.stage3_3_branches_0_1_2(stage3_3_branches_0_1_1)
        stage3_3_branches_1_0_conv1_0 = self.stage3_3_branches_1_0_conv1_0(stage3_2_relu_cbrs_1_0_2)
        stage3_3_branches_1_0_conv1_1 = self.stage3_3_branches_1_0_conv1_1(stage3_3_branches_1_0_conv1_0)
        stage3_3_branches_1_0_conv1_2 = self.stage3_3_branches_1_0_conv1_2(stage3_3_branches_1_0_conv1_1)
        stage3_3_branches_1_0_conv1_3 = self.stage3_3_branches_1_0_conv1_3(stage3_3_branches_1_0_conv1_2)
        stage3_3_branches_1_0_conv1_4 = self.stage3_3_branches_1_0_conv1_4(stage3_3_branches_1_0_conv1_3)
        add_36 = torch.add(stage3_3_branches_1_0_conv1_4, stage3_2_relu_cbrs_1_0_2)
        stage3_3_branches_1_1_0 = self.stage3_3_branches_1_1_0(add_36)
        stage3_3_branches_1_1_1 = self.stage3_3_branches_1_1_1(stage3_3_branches_1_1_0)
        stage3_3_branches_1_1_2 = self.stage3_3_branches_1_1_2(stage3_3_branches_1_1_1)
        stage3_3_branches_2_0_conv1_0 = self.stage3_3_branches_2_0_conv1_0(stage3_2_relu_cbrs_2_0_2)
        stage3_3_branches_2_0_conv1_1 = self.stage3_3_branches_2_0_conv1_1(stage3_3_branches_2_0_conv1_0)
        stage3_3_branches_2_0_conv1_2 = self.stage3_3_branches_2_0_conv1_2(stage3_3_branches_2_0_conv1_1)
        stage3_3_branches_2_0_conv1_3 = self.stage3_3_branches_2_0_conv1_3(stage3_3_branches_2_0_conv1_2)
        stage3_3_branches_2_0_conv1_4 = self.stage3_3_branches_2_0_conv1_4(stage3_3_branches_2_0_conv1_3)
        add_37 = torch.add(stage3_3_branches_2_0_conv1_4, stage3_2_relu_cbrs_2_0_2)
        stage3_3_branches_2_1_0 = self.stage3_3_branches_2_1_0(add_37)
        stage3_3_branches_2_1_1 = self.stage3_3_branches_2_1_1(stage3_3_branches_2_1_0)
        stage3_3_branches_2_1_2 = self.stage3_3_branches_2_1_2(stage3_3_branches_2_1_1)
        stage3_3_fuse_layers_0_1_0 = self.stage3_3_fuse_layers_0_1_0(stage3_3_branches_1_1_2)
        stage3_3_fuse_layers_0_1_1 = self.stage3_3_fuse_layers_0_1_1(stage3_3_fuse_layers_0_1_0)
        stage3_3_fuse_layers_0_1_2 = self.stage3_3_fuse_layers_0_1_2(stage3_3_fuse_layers_0_1_1)
        add_38 = torch.add(stage3_3_branches_0_1_2, stage3_3_fuse_layers_0_1_2)
        stage3_3_fuse_layers_0_2_0 = self.stage3_3_fuse_layers_0_2_0(stage3_3_branches_2_1_2)
        stage3_3_fuse_layers_0_2_1 = self.stage3_3_fuse_layers_0_2_1(stage3_3_fuse_layers_0_2_0)
        stage3_3_fuse_layers_0_2_2 = self.stage3_3_fuse_layers_0_2_2(stage3_3_fuse_layers_0_2_1)
        add_39 = torch.add(add_38, stage3_3_fuse_layers_0_2_2)
        stage3_3_relu_cbrs_0_0_0 = self.stage3_3_relu_cbrs_0_0_0(add_39)
        stage3_3_relu_cbrs_0_0_1 = self.stage3_3_relu_cbrs_0_0_1(stage3_3_relu_cbrs_0_0_0)
        stage3_3_relu_cbrs_0_0_2 = self.stage3_3_relu_cbrs_0_0_2(stage3_3_relu_cbrs_0_0_1)
        final_layers_0 = self.final_layers_0(stage3_3_relu_cbrs_0_0_2)
        cat_1 = torch.cat([stage3_3_relu_cbrs_0_0_2, final_layers_0], 1)
        deconv_layers_0_0_0 = self.deconv_layers_0_0_0(cat_1)
        deconv_layers_0_0_1 = self.deconv_layers_0_0_1(deconv_layers_0_0_0)
        deconv_layers_0_0_2 = self.deconv_layers_0_0_2(deconv_layers_0_0_1)
        deconv_layers_0_0_3 = self.deconv_layers_0_0_3(deconv_layers_0_0_2)
        deconv_layers_0_1_0_conv1_0 = self.deconv_layers_0_1_0_conv1_0(deconv_layers_0_0_3)
        deconv_layers_0_1_0_conv1_1 = self.deconv_layers_0_1_0_conv1_1(deconv_layers_0_1_0_conv1_0)
        deconv_layers_0_1_0_conv1_2 = self.deconv_layers_0_1_0_conv1_2(deconv_layers_0_1_0_conv1_1)
        deconv_layers_0_1_0_conv1_3 = self.deconv_layers_0_1_0_conv1_3(deconv_layers_0_1_0_conv1_2)
        deconv_layers_0_1_0_conv1_4 = self.deconv_layers_0_1_0_conv1_4(deconv_layers_0_1_0_conv1_3)
        deconv_layers_0_1_0_relu1 = self.deconv_layers_0_1_0_relu1(deconv_layers_0_1_0_conv1_4)
        deconv_layers_0_1_0_conv2_0 = self.deconv_layers_0_1_0_conv2_0(deconv_layers_0_1_0_relu1)
        deconv_layers_0_1_0_conv2_1 = self.deconv_layers_0_1_0_conv2_1(deconv_layers_0_1_0_conv2_0)
        deconv_layers_0_1_0_conv2_2 = self.deconv_layers_0_1_0_conv2_2(deconv_layers_0_1_0_conv2_1)
        deconv_layers_0_1_0_conv2_3 = self.deconv_layers_0_1_0_conv2_3(deconv_layers_0_1_0_conv2_2)
        deconv_layers_0_1_0_conv2_4 = self.deconv_layers_0_1_0_conv2_4(deconv_layers_0_1_0_conv2_3)
        add_40 = torch.add(deconv_layers_0_1_0_conv2_4, deconv_layers_0_0_3)
        final_layers_1 = self.final_layers_1(add_40)
        return final_layers_0, final_layers_1


if __name__ == "__main__":
    model = hrnet125()

    model.eval()
    model.cpu()

    dummy_input_0 = torch.ones((1, 3, 224, 224), dtype=torch.float32)

    output = model(dummy_input_0)
    print(output)