1D perceptron : RuntimeError: mat1 and mat2 shapes cannot be multiplied (1x3 and 1x1)

Question

This is part of my code, a simple 1d and 2 layer perceptron, and I want to implement minibatch.

class Net(nn.Module): ## nn.Module class is used
    def __init__(self):
        super(Net, self).__init__()                      
        self.fc1 = nn.Linear(1,1,bias=False)  # in dim, out dim
    def forward(self, x):
        x = self.fc1(x)
        return x

torch_dataset = Data.TensorDataset(X, Y)
Batch_size = 3
loader = Data.DataLoader(
    dataset=torch_dataset,
    batch_size=Batch_size,
    shuffle=True,
    num_workers=2,
)

if __name__ == '__main__':
    for epoch in range(20):  # 0 - 19
         for i, current_data in enumerate(loader):
             X, Y = current_data
             outputs = net(X)
             loss = criterion(outputs, Y)
             optimizer.zero_grad()
             loss.backward()  # lecture4,15min
             optimizer.step()  ## This line is equivalent to "W = W - lr* W.grad"
             print("Epoch {} - loss: {}".format(epoch, loss))

But it shows multiples errors and this.

RuntimeError: mat1 and mat2 shapes cannot be multiplied (1x3 and 1x1)

What am I doing wrong?