You are viewing a plain text version of this content. The canonical link for it is here.

Posted to commits@tvm.apache.org by GitBox <gi...@apache.org> on 2020/09/15 09:51:07 UTC

[GitHub] [incubator-tvm] wjliu1998 opened a new issue #6476: [TOPI] High dimension matrix multiplication not supported

wjliu1998 opened a new issue #6476:
URL: https://github.com/apache/incubator-tvm/issues/6476


   When run the code 
   `input_tensor_placeholder = te.placeholder((224,224,3), name="input_tensor_placeholder")
   m_tensor_placeholder = te.placeholder((3,3), name="m_tensor_placeholder")
   mid_result = topi.matmul(input_tensor_placeholder, m_tensor_placeholder)`
   error will occur:
   `TVMError: Check failed: ndim() == indices.size() (3 vs. 2) : Tensor dimension mismatch in readndim = 3, indices.size=2`
   It seems that in `topi.matmul`, the dimension of two tensors will be the same， which is different from the `torch.matmul`.
   
   Is there any other way to execute high dimension matrix multiplication? Or is it just a missing feature?


----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

For queries about this service, please contact Infrastructure at:
users@infra.apache.org