Library issue with DLRM v3

[FeelingFatigued]

Hi,

Thank you for the great work! I am trying to run DLRM v3, but it seems to require the Hammer library as shown below:

torch.ops.load_library("//deeplearning/fbgemm/fbgemm_gpu:sparse_ops")
torch.ops.load_library("//deeplearning/fbgemm/fbgemm_gpu:sparse_ops_cpu")
torch.ops.load_library("//hammer/oss/generative_recommenders/ops/cpp:cpp_ops")

from hammer.fx_utils import fx_unwrap_optional_tensor

Is it possible to run DLRM v3 without this library, or could you help fix these issues?

I also saw a comment stating that DLRM v3 is still under development. Does this mean it's currently not possible to run the model?

Thank you in advance for the help!