masahi commented on a change in pull request #8985:
URL: https://github.com/apache/tvm/pull/8985#discussion_r709112076
##########
File path: python/tvm/relay/op/strategy/cuda.py
##########
@@ -1210,3 +1210,16 @@ def invert_permutation_strategy_cuda(attrs, inputs,
out_type, target):
name="invert_permutation.cuda",
)
return strategy
+
+
+@einsum_strategy.register(["cuda", "gpu"])
+def einsum_strategy_cuda(attrs, inputs, out_type, target):
+ """einsum cuda strategy"""
+ strategy = _op.OpStrategy()
+ # TODO: Add cuda-specific op implementation for einsum
+ strategy.add_implementation(
Review comment:
This would lead to an error if you try to run it on cuda right? It is
better to remove this strategy until we have a CUDA schedule ready. The error
message would be clearer than the one from incorrect scheduling.
--
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]
