Correctly changing precision in DLRM

Hello,

I would like to change the precision used in DLRM inference. I observed that the default datatype used in the original code is "float32", in the [create_emb()](https://github.com/facebookresearch/dlrm/blob/9830654a2b33f0ff11900fb1622e095a23eb2cab/dlrm_s_pytorch.py#L284) method. Is changing the dtype to any of the [supported values](https://pytorch.org/docs/stable/tensor_attributes.html) enough to correctly change the precision?

I believe that the generated embedding table would be directly used in the [apply_emb()](https://github.com/facebookresearch/dlrm/blob/9830654a2b33f0ff11900fb1622e095a23eb2cab/dlrm_s_pytorch.py#L456) function.

Thanks for the help.
Rishabh Jain