M2MRF配置信息

已于 2023-09-22 10:36:36 修改
阅读量63 收藏
点赞数
文章标签： 深度学习 python pytorch
于 2023-09-22 07:15:05 首次发布
版权声明：本文为博主原创文章，遵循 CC 4.0 BY-SA 版权协议，转载请附上原文出处链接和本声明。
本文链接：https://blog.csdn.net/weixin_44883789/article/details/133146900
版权

M2MRF配置信息

摘要：由于文章内容限制，【M2MRF代码解读】文章放不下那么多文本，便在此文章中展示所有配置信息。

1. env_info环境信息

env_info：环境信息如下

sys.platform: linux Python: 3.7.12 | packaged by conda-forge |
(default, Oct 26 2021, 06:08:21) [GCC 9.4.0] CUDA available: True GPU
0: NVIDIA TITAN Xp CUDA_HOME: /usr/local/cuda-10.2: GCC: gcc (GCC)
5.2.0 PyTorch: 1.6.0 PyTorch compiling details: PyTorch built with:

  • GCC 7.3
  • C++ Version: 201402
  • Intel® oneAPI Math Kernel Library Version 2021.4-Product Build 20210904 for Intel® 64 architecture applications
  • Intel® MKL-DNN v1.5.0 (Git Hash e2ac1fac44c5078ca927cb9b90e1b3066a0b2ed0)
  • OpenMP 201511 (a.k.a. OpenMP 4.5)
  • NNPACK is enabled
  • CPU capability usage: AVX2
  • CUDA Runtime 10.2
  • NVCC architecture flags: -gencode;arch=compute_37,code=sm_37;-gencode;arch=compute_50,code=sm_50;-gencode;arch=compute_60,code=sm_60;-gencode;arch=compute_61,code=sm_61;-gencode;arch=compute_70,code=sm_70;-gencode;arch=compute_75,code=sm_75;-gencode;arch=compute_37,code=compute_37
  • CuDNN 7.6.5
  • Magma 2.5.2
  • Build settings: BLAS=MKL, BUILD_TYPE=Release, CXX_FLAGS= -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DUSE_VULKAN_WRAPPER -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow, PERF_WITH_AVX=1, PERF_WITH_AVX2=1, PERF_WITH_AVX512=1, USE_CUDA=ON,
    USE_EXCEPTION_PTR=1, USE_GFLAGS=OFF, USE_GLOG=OFF, USE_MKL=ON,
    USE_MKLDNN=ON, USE_MPI=OFF, USE_NCCL=ON, USE_NNPACK=ON, USE_OPENMP=ON,
    USE_STATIC_DISPATCH=OFF,

TorchVision: 0.7.0 OpenCV: 4.7.0 MMCV: 1.2.0 MMCV Compiler: GCC 7.3
MMCV CUDA Compiler: 10.2 MMSegmentation: 0.8.0+

2. 模型信息：model

这部分是cfg.pretty_text的内容：

EncoderDecoder( (backbone): HRNet_M2MRF_C(

(conv1): Conv2d(3, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1), bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
(layer1): Sequential(
  (0): Bottleneck(
    (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU(inplace=True)
    (downsample): Sequential(
      (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
      (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    )
  )
  (1): Bottleneck(
    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU(inplace=True)
  )
  (2): Bottleneck(
    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU(inplace=True)
  )
  (3): Bottleneck(
    (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (relu): ReLU(inplace=True)
  )
)
(transition1): ModuleList(
  (0): Sequential(
    (0): Conv2d(256, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
    (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
  )
  (1): Sequential(
    (0): Sequential(
      (0): M2MRF(
        (sample_encode_conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1))
        (sample): M2MRF_Module(
          (sample_fc): Conv1d(4096, 64, kernel_size=(1,), stride=(1,))
          (sample_fc1): Conv1d(64, 1024, kernel_size=(1,), stride=(1,))
        )
        (sample_decode_conv): Conv2d(64, 96, kernel_size=(1, 1), stride=(1, 1))
      )
      (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace=True)
    )
  )
)
(stage2): Sequential(
  (0): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
      )
    )
    (relu): ReLU()
  )
)
(transition2): ModuleList(
  (0): None
  (1): None
  (2): Sequential(
    (0): Sequential(
      (0): M2MRF(
        (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
        (sample): M2MRF_Module(
          (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
          (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
        )
        (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
      )
      (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace=True)
    )
  )
)
(stage3): Sequential(
  (0): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
      )
    )
    (relu): ReLU()
  )
  (1): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
      )
    )
    (relu): ReLU()
  )
  (2): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
      )
    )
    (relu): ReLU()
  )
  (3): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
      )
    )
    (relu): ReLU()
  )
)
(transition3): ModuleList(
  (0): None
  (1): None
  (2): None
  (3): Sequential(
    (0): Sequential(
      (0): M2MRF(
        (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
        (sample): M2MRF_Module(
          (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
          (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
        )
        (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
      )
      (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
      (2): ReLU(inplace=True)
    )
  )
)
(stage4): Sequential(
  (0): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (3): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
        (3): Sequential(
          (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (3): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (2): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (3): None
      )
    )
    (relu): ReLU()
  )
  (1): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (3): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
        (3): Sequential(
          (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (3): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (2): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (3): None
      )
    )
    (relu): ReLU()
  )
  (2): HRModule_M2MRF(
    (branches): ModuleList(
      (0): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (1): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (2): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
      (3): Sequential(
        (0): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (1): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (2): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
        (3): BasicBlock(
          (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
          (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (relu): ReLU(inplace=True)
        )
      )
    )
    (fuse_layers): ModuleList(
      (0): ModuleList(
        (0): None
        (1): Sequential(
          (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (2): Sequential(
          (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (1): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): None
        (2): Sequential(
          (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
        (3): Sequential(
          (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (2): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): None
        (3): Sequential(
          (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
          )
        )
      )
      (3): ModuleList(
        (0): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (2): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (1): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
            (2): ReLU()
          )
          (1): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (2): Sequential(
          (0): Sequential(
            (0): M2MRF(
              (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
              (sample): M2MRF_Module(
                (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
                (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
              )
              (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
            )
            (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          )
        )
        (3): None
      )
    )
    (relu): ReLU()
  )
)   )   (decode_head): FCNHead(
input_transform=resize_concat, ignore_index=255, align_corners=False
(loss_decode): BinaryLoss()
(conv_seg): Conv2d(720, 4, kernel_size=(1, 1), stride=(1, 1))
(convs): Sequential(
  (0): ConvModule(
    (conv): Conv2d(720, 720, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (bn): BatchNorm2d(720, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (activate): ReLU(inplace=True)
  )
)   ) )

3. pipeline

class CustomDataset中的pipeline参数：

[{‘type’: ‘LoadImageFromFile’},
{‘type’: ‘LoadAnnotations’},
{‘type’: ‘Resize’, ‘img_scale’: (512, 512), ‘ratio_range’: (0.5, 2.0)},
{‘type’: ‘RandomCrop’, ‘crop_size’: (512, 512), ‘cat_max_ratio’:0.75},
{‘type’: ‘RandomFlip’, ‘flip_ratio’: 0},
{‘type’: ‘PhotoMetricDistortion’},
{‘type’: ‘Normalize’, ‘mean’: [116.513, 56.437, 16.309], ‘std’: [80.206, 41.232, 13.293], ‘to_rgb’: True},
{‘type’: ‘Pad’, ‘size’: (512, 512), ‘pad_val’: 0, ‘seg_pad_val’: 0},
{‘type’: ‘DefaultFormatBundle’},
{‘type’: ‘Collect’, ‘keys’: [‘img’, ‘gt_semantic_seg’]}]

4. self.pipeline

Compose(
LoadImageFromFile(to_float32=False,color_type=‘color’,imdecode_backend=‘cv2’)
LoadAnnotations(reduce_zero_label=False,imdecode_backend=‘pillow’)
Resize(img_scale=[(512, 512)], multiscale_mode=range, ratio_range=(0.5, 2.0), keep_ratio=True)
RandomCrop(crop_size=(512, 512))
RandomFlip(flip_ratio=0)
PhotoMetricDistortion(brightness_delta=32, contrast_range=(0.5, 1.5), saturation_range=(0.5, 1.5), hue_delta=18)
Normalize(mean=[116.513 56.437 16.309], std=[80.206 41.232 13.293], to_rgb=True)
Pad(size=(512, 512), size_divisor=None, pad_val=0)
DefaultFormatBundle
Collect(keys=[‘img’, ‘gt_semantic_seg’], meta_keys=(‘filename’, ‘ori_filename’, ‘ori_shape’, ‘img_shape’, ‘pad_shape’, ‘scale_factor’,
‘flip’, ‘flip_direction’, ‘img_norm_cfg’)) )

5. HRNet_M2MRF_C

HRNet_M2MRF_C(
(conv1): Conv2d(3, 64, kernel_size=(3, 3), stride=(2,2), padding=(1, 1), bias=False)
(bn1): BatchNorm2d(64, eps=1e-05,momentum=0.1, affine=True, track_running_stats=True)
(conv2):Conv2d(64, 64, kernel_size=(3, 3), stride=(2, 2), padding=(1, 1),bias=False)
(bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1,affine=True, track_running_stats=True)
(relu): ReLU(inplace=True)
 
(layer1): Sequential(

(0): Bottleneck(
  (conv1): Conv2d(64, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
  (downsample): Sequential(
    (0): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
    (1): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  )
)
(1): Bottleneck(
  (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
)
(2): Bottleneck(
  (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
)
(3): Bottleneck(
  (conv1): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn1): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv2): Conv2d(64, 64, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (bn2): BatchNorm2d(64, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (conv3): Conv2d(64, 256, kernel_size=(1, 1), stride=(1, 1), bias=False)
  (bn3): BatchNorm2d(256, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (relu): ReLU(inplace=True)
)   )   (transition1): ModuleList(
(0): Sequential(
  (0): Conv2d(256, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
  (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
  (2): ReLU(inplace=True)
)
(1): Sequential(
  (0): Sequential(
    (0): M2MRF(
      (sample_encode_conv): Conv2d(256, 64, kernel_size=(1, 1), stride=(1, 1))
      (sample): M2MRF_Module(
        (sample_fc): Conv1d(4096, 64, kernel_size=(1,), stride=(1,))
        (sample_fc1): Conv1d(64, 1024, kernel_size=(1,), stride=(1,))
      )
      (sample_decode_conv): Conv2d(64, 96, kernel_size=(1, 1), stride=(1, 1))
    )
    (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
  )
)   )   (stage2): Sequential(
(0): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
    )
  )
  (relu): ReLU()
)   )   (transition2): ModuleList(
(0): None
(1): None
(2): Sequential(
  (0): Sequential(
    (0): M2MRF(
      (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
      (sample): M2MRF_Module(
        (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
        (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
      )
      (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
    )
    (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
  )
)   )   (stage3): Sequential(
(0): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
    )
  )
  (relu): ReLU()
)
(1): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
    )
  )
  (relu): ReLU()
)
(2): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
    )
  )
  (relu): ReLU()
)
(3): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
    )
  )
  (relu): ReLU()
)   )   (transition3): ModuleList(
(0): None
(1): None
(2): None
(3): Sequential(
  (0): Sequential(
    (0): M2MRF(
      (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
      (sample): M2MRF_Module(
        (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
        (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
      )
      (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
    )
    (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
    (2): ReLU(inplace=True)
  )
)   )   (stage4): Sequential(
(0): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (3): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
      (3): Sequential(
        (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (3): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (2): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (3): None
    )
  )
  (relu): ReLU()
)
(1): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (3): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
      (3): Sequential(
        (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (3): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (2): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (3): None
    )
  )
  (relu): ReLU()
)
(2): HRModule_M2MRF(
  (branches): ModuleList(
    (0): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(48, 48, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (1): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(96, 96, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (2): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(192, 192, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
    (3): Sequential(
      (0): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (1): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (2): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
      (3): BasicBlock(
        (conv1): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (conv2): Conv2d(384, 384, kernel_size=(3, 3), stride=(1, 1), padding=(1, 1), bias=False)
        (bn2): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (relu): ReLU(inplace=True)
      )
    )
  )
  (fuse_layers): ModuleList(
    (0): ModuleList(
      (0): None
      (1): Sequential(
        (0): Conv2d(96, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 3072, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (2): Sequential(
        (0): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 48, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(12, 49152, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (1): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): None
      (2): Sequential(
        (0): Conv2d(192, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 6144, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
      (3): Sequential(
        (0): Conv2d(384, 96, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(24, 24576, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (2): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 192, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): None
      (3): Sequential(
        (0): Conv2d(384, 192, kernel_size=(1, 1), stride=(1, 1), bias=False)
        (1): BatchNorm2d(192, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        (2): M2MRF(
          (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
          (sample): M2MRF_Module(
            (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
            (sample_fc1): Conv1d(48, 12288, kernel_size=(1,), stride=(1,))
          )
          (sample_decode_conv): Conv2d(48, 192, kernel_size=(1, 1), stride=(1, 1))
        )
      )
    )
    (3): ModuleList(
      (0): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 48, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(48, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (2): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(48, 12, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(768, 12, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(12, 192, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(12, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (1): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 96, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(96, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
          (2): ReLU()
        )
        (1): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(96, 24, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(1536, 24, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(24, 384, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(24, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (2): Sequential(
        (0): Sequential(
          (0): M2MRF(
            (sample_encode_conv): Conv2d(192, 48, kernel_size=(1, 1), stride=(1, 1))
            (sample): M2MRF_Module(
              (sample_fc): Conv1d(3072, 48, kernel_size=(1,), stride=(1,))
              (sample_fc1): Conv1d(48, 768, kernel_size=(1,), stride=(1,))
            )
            (sample_decode_conv): Conv2d(48, 384, kernel_size=(1, 1), stride=(1, 1))
          )
          (1): BatchNorm2d(384, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True)
        )
      )
      (3): None
    )
  )
  (relu): ReLU()
)   ) )
Cpdr
关注
M2MRF代码说明及结果展示
weixin_44883789的博客
08-06 147
对四种病变类型进行标注（在同一张图片上显示），保存在annotations文件夹中。（1）四种病变标签：（保存在visual/baseline文件夹）其中visual（蓝色选中）文件夹中也是。：对图片进行数据增强，在。中对每一张图片分别进行。
【代码复现】M2MRF采样算子代码实现流程
weixin_44883789的博客
09-21 405
对M2MRF分割模型代码逐行解读，该模型使用mmsegmentation架构，模型的创新点是改进采样算子，使用HRnet作为分割网络。
mmDetection源码分析（2）：训练与配置文件
cuihaoren01的专栏
12-03 1730
训练函数调用 训练脚本 python tools/train.py configs/faster_rcnn_r50_fpn_1x.py 训练可选项： –work_dir：模型checkpoint以及训练log文件的输出目录，若在脚本中不设置，则为config/*.py中work_dir中的路径。 –resume_from：指定在某个checkpoint的基础上继续训练，若在脚本中不设置，则为co...
【Python:Pycharm】mmSegmentation语义分割框架教程
weixin_44883789的博客
07-07 6852
MMSegmentation是openmmlab项目下开源的图像语义分割框架，目前支持pytorch，由于其拥有pipeline加速，完善的数据增强体系，完善的模型库，作为大数据语义分割训练及测试的代码框架是再好不过了。
【读点论文】WEAKLY-SUPERVISED TEXT INSTANCE SEGMENTATION 通过文本转录信息就能实现字符实例分割，看示例效果可以，期待源码开放解析
最新发布
白水空空
09-11 1042
由于分割模块的伪标签来自识别器生成的注意力图，如果识别器对文本实例位置的认识有错误，细化模块将不会产生有效的伪标签，如图 6 所示。此外，正如中间一行所示，一个有趣的观察是，对于像“C”、“O”和“D”这样的文本实例，它们初始注意力的细化有时会被它们自己的空心结构误导（这是不可避免的），并且模型需要长时间的联合训练才能摆脱这种空心陷阱。因此，当识别模块和分割模块进行联合训练时，识别器生成的注意图逐渐学会更接近相应文本实例的形状（如图 7（c）和图 7（d）所示），从而为分割模块提供质量更好的伪标签。
android编译 cryptopp_GDAL-Android交叉编译armV7
weixin_42414213的博客
01-12 288
前言仅考虑GDAL-OGR部分已经ArmV7平台准备NDK./make-standalone-toolchain.sh --toolchain=arm-linux-androideabi-4.9 --arch=arm --platform=android-15 --system=windows-x86_64 --install-dir=/cygdrive/d/android/android-to...
如何创造一个能和你对话的语音AI？
阿里技术
06-23 5786
过去十年，语音AI从实验室走向应用，语音搜索、交互早已融入日常。本文将带你一览达摩院语音AI技术创新全景，一起感受能听、会说、懂你的语音AI。
JAVA 二维码图片生成和base64位加密解密
weixin_37530941的博客
08-08 7595
二维码生成工具 带logo二维码 二维码base64加密解密 二维码识别
达摩院技术创新全景｜懂你的语音AI
阿里云技术
07-04 599
过去十年，语音AI从实验室走向应用，语音搜索、交互早已融入日常。本文将带你一览达摩院语音AI技术创新全景，一起感受能听、会说、懂你的语音AI。
IRINA：具有鲁棒性的虹膜识别算法及其在不准确分割数据中的应用
1IRINA：不准确分割数据中的虹膜识别（偶数）HugoProen C. Nev es IT-葡萄牙贝拉内政大学电信学院{hugomcp，jcneves}@ di.ubi.pt摘要当前虹膜识别系统...这样的信息使得能够推断图像之间的自由形式的变形场（2D配准向量
学习图正则化的超分辨率引导方法
学习的图形潜力使利用丰富的控制成为可能，感受野源引导数据保真度从引导图像的文本信息，而前，... ...这是什么？图体系结构内的显式图形优化保证了高分辨率目标对低分辨率目标的严格保真度了解到正则化分辨率来源
pycharm调试知识， 线程进程与深度学习
小泽爱刷题的博客
09-10 1196
多线程在深度学习中主要用于I/O 密集型任务，如数据加载和预处理，以防止数据成为训练过程中的瓶颈。多进程常用于克服 GIL 的限制，处理 CPU 密集型任务，以及在分布式训练中并行化模型训练。GPU 并行计算是深度学习的核心，主要负责大规模矩阵运算和模型训练，并行性是由 GPU 硬件架构提供的，而不是通过 Python 的多线程机制。因此，深度学习中多线程的应用主要集中在数据加载、推理服务和辅助任务，而核心计算则由 GPU 完成并行化处理。多进程。
【Python机器学习】循环神经网络（RNN）——审察模型内部情况
weixin_39407597的博客
09-10 642
Keras附带了一些工具，比如model.summary()，用于审察模型内部情况。随着模型变得越来越复杂，我们需要经常使用model.summary()，否则在调整超参数时跟踪模型内部的内容的变化情况会变得非常费力。
深度学习：怎么看pth文件的参数
weixin_59815450的博客
09-07 513
你需要有创建这个.pth文件时所用的模型的代码。这意味着你需要有模型的类定义和架构。
LSTM处理时序数据：深入解析与实战
异构算力老群群（在读985计算机博士生）的技术博客
09-10 1088
LSTM，全称为Long Short-Term Memory，是一种特殊的RNN结构，旨在解决传统RNN在处理长序列时容易出现的梯度消失或爆炸问题。LSTM通过引入“门”机制和细胞状态，使得网络能够更好地保留长期记忆。
如何将卷积神经网络（CNN）应用于医学图像分析：从分类到分割和检测的实用指南
loveting99的博客
09-07 1025
U-Net：U-Net是一种用于生物医学图像分割的卷积网络。它的关键特点是具有对称的编码器-解码器结构和跳跃连接（skip connections），可以更好地保留和利用图像的空间信息。DeepLab：DeepLab使用空洞卷积（dilated convolutions）来扩展卷积核的感受野，同时保持图像的分辨率，并结合条件随机场（CRFs）进一步精细化分割结果。：YOLO通过将图像划分为网格，并在每个网格中预测边界框和类别，具有实时检测的优势。
如何理解深度学习的训练过程
qq_61033357的博客
09-11 307
这是模型在训练过程中通过数据自动学习的参数，主要是卷积层的权重和偏置。比如卷积核内部的权重（Conv2d中的weights和bias）会随着训练数据的输入和误差的反馈不断更新。以yolov5为例子，训练的目的是把一组输入猫狗图像放到神经网络中，得到一个输出模型，这个模型下次可以直接用来识别哪个是猫，哪个是狗。这些参数是在反向传播过程中由优化器（如Adam、SGD等）自动调整的，目的是使模型更好地拟合训练数据。步长（stride）：决定卷积操作在输入特征图上的移动幅度，影响输出的特征图大小。
对抗性EM用于变分深度学习：在低剂量PET和低剂量CT中的半监督图像质量增强应用|文献速递--Transformer架构在医学影像分析中的应用
weixin_38594676的博客
09-11 919
在PET中 𝛾 = 0.5）：(i) N2N+UNet，(ii) N2N+UNet+Adv，在N2N+UNet中加入对抗学习，(iii) N2N+UNet+DistLat，在N2N+UNet的潜在空间中加入基于EM的分布建模，(iv) N2N+UNet+DistOut，在N2N+UNet的解码器输出处加入GG分布建模，(v) ss-NonAdv-EM，将DistLat和DistOut都加入N2N+UNet，(vi) ssAdvEM，在ss-NonAdv-EM中加入对抗学习。图 2. 结果：定量分析。
评论
添加红包

请填写红包祝福语或标题

个

红包个数最小为10个

元

红包金额最低5元

当前余额3.43元 前往充值 >
需支付：10.00元
成就一亿技术人!
领取后你会自动成为博主和红包主的粉丝 规则
hope_wisdom
发出的红包

打赏作者

Cpdr

你的鼓励将是我创作的最大动力

¥1 ¥2 ¥4 ¥6 ¥10 ¥20
扫码支付：¥1
获取中
扫码支付

您的余额不足，请更换扫码支付或充值

打赏作者

实付元
使用余额支付
点击重新获取
扫码支付
钱包余额 0

抵扣说明：

1.余额是钱包充值的虚拟货币，按照1:1的比例进行支付金额的抵扣。
2.余额无法直接购买下载，可以购买VIP、付费专栏及课程。

余额充值