Given groups=1, weight of size [16, 3, 3, 3], expected input[4, 4, 2048, 1536] to have 3 channels, but got 4 channels instead

[rebecca0011]

I only downloaded the 00008 data as a training and validation set due to network problems, but got this error

/home/rc/anaconda3/envs/DL/bin/python3.7 /home/rc/StereoMatching/SMD-Nets/apps/train.py --dataroot /home/rc/StereoMatching/SMD-Nets/Dataset/UnrealStereo4K_00008 --checkpoints_path /home/rc/StereoMatching/SMD-Nets/checkpoints --training_file /home/rc/StereoMatching/SMD-Nets/filenames/train2.txt --testing_file /home/rc/StereoMatching/SMD-Nets/filenames/test2.txt --results_path /home/rc/StereoMatching/SMD-Nets/output --name train --mode passive --batch_size 2 --num_epoch 1 --learning_rate 1e-4 --gamma 0.1 --crop_height 2048 --crop_width 1536 --num_sample_inout 5000 --aspect_ratio 1. --sampling dda --output_representation bimodal --backbone HSMNet
train data size:  92
test data size:  10
/home/rc/StereoMatching/SMD-Nets/checkpoints/train
Start Training
Traceback (most recent call last):
  File "/home/rc/StereoMatching/SMD-Nets/apps/train.py", line 196, in <module>
    train(opt)
  File "/home/rc/StereoMatching/SMD-Nets/apps/train.py", line 113, in train
    errors = net.forward(left, right, sample, labels=labels)
  File "/home/rc/StereoMatching/SMD-Nets/lib/model/SMDHead.py", line 108, in forward
    self.filter(left, right)
  File "/home/rc/StereoMatching/SMD-Nets/lib/model/SMDHead.py", line 23, in filter
    self.feat_list = self.stereo_network(left, right)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/rc/StereoMatching/SMD-Nets/lib/backbone/HSMNet/hsm.py", line 60, in forward
    conv4, conv3, conv2, conv1, enc0, enc1, enc2, enc3 = self.feature_extraction(torch.cat([left, right], 0))
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/rc/StereoMatching/SMD-Nets/lib/backbone/HSMNet/utils.py", line 80, in forward
    conv1 = self.convbnrelu1_1(x)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/rc/StereoMatching/SMD-Nets/lib/backbone/HSMNet/utils.py", line 154, in forward
    outputs = self.cbr_unit(inputs)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/container.py", line 119, in forward
    input = module(input)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/module.py", line 889, in _call_impl
    result = self.forward(*input, **kwargs)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 399, in forward
    return self._conv_forward(input, self.weight, self.bias)
  File "/home/rc/anaconda3/envs/DL/lib/python3.7/site-packages/torch/nn/modules/conv.py", line 396, in _conv_forward
    self.padding, self.dilation, self.groups)
RuntimeError: Given groups=1, weight of size [16, 3, 3, 3], expected input[4, 4, 2048, 1536] to have 3 channels, but got 4 channels instead

Process finished with exit code 1

The parameters I use are set like this
--dataroot /home/rc/StereoMatching/SMD-Nets/Dataset/UnrealStereo4K_00008 --checkpoints_path /home/rc/StereoMatching/SMD-Nets/checkpoints --training_file /home/rc/StereoMatching/SMD-Nets/filenames/train2.txt --testing_file /home/rc/StereoMatching/SMD-Nets/filenames/test2.txt --results_path /home/rc/StereoMatching/SMD-Nets/output --name train --mode passive --batch_size 2 --num_epoch 1 --learning_rate 1e-4 --gamma 0.1 --crop_height 2048 --crop_width 1536 --num_sample_inout 5000 --aspect_ratio 1. --sampling dda --output_representation "bimodal" --backbone HSMNet