RuntimeError: size mismatch, m1: [32 x 5184], m2: [400 x 120] at /opt/conda/conda-bld/pytorch_1570910687230/work/aten/src/THC/generic/THCTensorMathBlas.cu:290

1

I don’t know whats wrong with my LeNet,I can not understand why it could be wrong

class LeNet(nn.Module):
    def __init__(self):
        super(LeNet,self).__init__()
        self.conv1=nn.Conv2d(3,6,5)
        self.conv2=nn.Conv2d(6,16,5)
        self.fc1=nn.Linear(16*5*5,120)
        self.fc2=nn.Linear(120,84)
        self.fc3=nn.Linear(84,48)
    def forward(self,x):
        out = F.relu(self.conv1(x))
        out = F.max_pool2d(out, 2)
        out = F.relu(self.conv2(out))
        out = F.max_pool2d(out, 2)
        out = out.view(out.size(0), -1)
        out = F.relu(self.fc1(out))
        out = F.relu(self.fc2(out))
        out = self.fc3(out)
        return x

Traceback (most recent call last):
  File "/home/xiu/Desktop/mediacompute/ImageRecognition/main1.py", line 97, in <module>
    main(config)
  File "/home/xiu/Desktop/mediacompute/ImageRecognition/main1.py", line 48, in main
    solver.train()
  File "/home/xiu/Desktop/mediacompute/ImageRecognition/solver.py", line 135, in train
    outputs =self.model(inputs)
  File "/home/xiu/anaconda3/envs/torch1.0/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/xiu/Desktop/mediacompute/ImageRecognition/Lenet.py", line 17, in forward
    out = F.relu(self.fc1(out))
  File "/home/xiu/anaconda3/envs/torch1.0/lib/python3.6/site-packages/torch/nn/modules/module.py", line 541, in __call__
    result = self.forward(*input, **kwargs)
  File "/home/xiu/anaconda3/envs/torch1.0/lib/python3.6/site-packages/torch/nn/modules/linear.py", line 87, in forward
    return F.linear(input, self.weight, self.bias)
  File "/home/xiu/anaconda3/envs/torch1.0/lib/python3.6/site-packages/torch/nn/functional.py", line 1370, in linear
    ret = torch.addmm(bias, input, weight.t())
RuntimeError: size mismatch, m1: [32 x 5184], m2: [400 x 120] at /opt/conda/conda-bld/pytorch_1570910687230/work/aten/src/THC/generic/THCTensorMathBlas.cu:290
2

It seems the flattened output of self.conv2 and F.max_pool seems to be [batch_size, 5184], while self.fc1 expects an input of [batch_size, 400].
Which input size are you using?

1 Like
3

Hy sir am getting same type of error and the error is

erro
erro1298×70 3.24 KB
i cannot be able to understand this that why it is occuring?
Or either i changed the num of workers and batch size is it due to that one? kindly guide me how to solve it.

4

And the relevent part of the code is

parser
parser960×448 16.4 KB

5

Could you post your model definition as well as the input shapes?
Please post the code by wrapping it into three backticks ``` instead of pictures, as this is easier to debug.

1 Like
6

hy sir actually i gave three features in my original model but am trying to update it in one feature but its giving this error wich i posted kindly guide me that how can i post model and training file so that you can guide me?

7

Moreover sir i cant be able to upload the model because here one can just upload the jpg kindly guide me how can i upload my model and the new one so that you may help me to debug it