Mousius commented on code in PR #12455:
URL: https://github.com/apache/tvm/pull/12455#discussion_r1001688225
##########
tests/python/contrib/test_arm_compute_lib/test_network.py:
##########
@@ -180,7 +183,12 @@ def get_model():
return mod, params, inputs
_build_and_run_network(
- *get_model(), device=device, tvm_ops=3, acl_partitions=30, atol=9,
rtol=0
+ *get_model(),
+ device=device,
+ tvm_ops=3,
+ acl_partitions=30,
+ atol=9,
+ rtol=0.01,
Review Comment:
I'm guessing as we now run slightly different schedules we get slightly
different answers, this aligns with the above `rtol` though.
--
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.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
