TylerADavis commented on a change in pull request #7134:
URL: https://github.com/apache/tvm/pull/7134#discussion_r546871956
##########
File path: tests/python/frontend/pytorch/test_forward.py
##########
@@ -1889,21 +1889,73 @@ def _get_default_vm_targets():
return [tgt for (tgt, _) in tvm.testing.enabled_targets()]
-def verify_script_model(pt_model, ishapes, targets):
+def verify_script_model(pt_model, ishapes, targets, idtype=None):
script_module = torch.jit.script(pt_model)
verify_model_vm(script_module, ishapes, targets=targets)
+ if idtype:
+ verify_model_vm(script_module, ishapes, idtype=idtype, targets=targets)
+ else:
+ verify_model_vm(script_module, ishapes, targets=targets)
+
Review comment:
Done. I've cleaned up `verify_script_model()` and changed the handling
of default arguments in `verify_model_vm()` so that passing in `None` will
result in a `torch.float` being used. Let me know if the new approach looks
good.
----------------------------------------------------------------
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:
[email protected]
