Transform the target image with the transform made in step 3 and add an extra batch dimension with `torch.unsqueeze(dim=0)` so our input image has shape `[batch_size, color_channels, height, width]`.\n", "7. Make a prediction on the image by passing it to the model ensurin...