ANSHUMAN87 commented on a change in pull request #7435:
URL: https://github.com/apache/tvm/pull/7435#discussion_r587517097
##########
File path: python/tvm/relay/op/strategy/generic.py
##########
@@ -799,6 +799,36 @@ def sparse_dense_padded_strategy(attrs, inputs, out_type,
target):
raise NotImplementedError("sparse_dense_padded is only implemented for
cuda")
+# sparse_add
+def wrap_compute_sparse_add(topi_compute):
+ """wrap sparse add topi compute"""
+
+ def _compute_sparse_add(attrs, inputs, out_type):
+ return [topi_compute(inputs[0], inputs[1], inputs[2], inputs[3])]
+
+ return _compute_sparse_add
+
+
+@override_native_generic_func("sparse_add_strategy")
+def sparse_add_strategy(attrs, inputs, out_type, target):
+ """sparse add generic strategy"""
+ logger.warning("sparse add is not optimized for this platform.")
Review comment:
@tkonolige : Sorry for late response was stuck in another . I think this
is general convention for generic strategy.
----------------------------------------------------------------
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]
