PyTorch: Conv2D Forward Pass

Implement a Conv2D forward pass using PyTorch's built-in torch.nn.functional.conv2d.

Signature: def conv2d_forward_backward(x, weight, bias, stride=1, padding=0)

• x: input tensor of shape (N, C_in, H, W)
• weight: filter tensor (C_out, C_in, kH, kW)
• bias: bias tensor (C_out,)
• stride: convolution stride (default 1)
• padding: zero-padding size (default 0)
• Returns: output of the convolution as a nested list

Use torch.nn.functional.conv2d for the forward pass. The output shape is (N, C_out, H_out, W_out) where H_out = (H + 2*padding - kH) // stride + 1.

Example:

x = [[[[1,1,1],[1,1,1],[1,1,1]]]]  # (1,1,3,3)
w = [[[[1,1],[1,1]]]]              # (1,1,2,2)
b = [0.0]
# Output shape: (1,1,2,2), each position sums 4 ones = 4.0