Add SLU models

[siddhu001]

[codecov-commenter]

Codecov Report

Merging #56 (03bbd39) into master (8b73019) will not change coverage. The diff coverage is n/a.

@@           Coverage Diff           @@
##           master      #56   +/-   ##
=======================================
  Coverage   50.76%   50.76%           
=======================================
  Files           2        2           
  Lines         390      390           
=======================================
  Hits          198      198           
  Misses        192      192           

:mega: We’re building smart automated test selection to slash your CI/CD build times. Learn more

[siddhu001]

@ftshijt Do you know how I can fix this issue?

[sw005320]

Are these models using s3prl? If so, the test should include it in the installation stage.

It is a little bit difficult decision as we want to make espnet_model_zoo installation as light as possible. But anyway, can you first try it?

[siddhu001]

They do use s3prl. Sure I will do that!

[sw005320]

Can you fix the conflict?

[siddhu001]

Sure! I will work on it now!

[siddhu001]

I think the earlier error due to ASR streaming model has been fixed now with the help of previously merged pull request - https://github.com/espnet/espnet_model_zoo/pull/57.

I have fixed the import ordering. I think after this pull request is merged - https://github.com/espnet/espnet/pull/3928, the test should hopefully pass.

[siddhu001]

The issue seems to be connected to https://github.com/pytorch/fairseq/issues/1400. There seems to be some issue with pytorch hub cache. I will try using workaround mentioned in this issue.