Concatenate inplace

Hi, it there anyway to concatenate tensors inplace to use memory efficiently?
Like Tensor:cat() in Torch7