giuseros commented on a change in pull request #6794:
URL: https://github.com/apache/incubator-tvm/pull/6794#discussion_r520667413
##########
File path: python/tvm/topi/x86/nn.py
##########
@@ -69,3 +69,33 @@ def schedule_softmax(outs):
s[exp].compute_at(s[softmax], fused_outer_axes)
return s
+
+
+def schedule_embed_grad(outs):
+ """Schedule for embed_grad
+
+ Parameters
+ ----------
+ outs: Array of Tensor
+ The computation graph description of embed_grad
+ in the format of an array of tensors.
+
+ Returns
+ -------
+ sch: Schedule
+ The computation schedule for the op.
+ """
+ s = te.create_schedule([outs[0].op])
+
+ vec_size = 8 # should autotune this, but we can't with hybrid script
Review comment:
May I ask why 8? I am just wondering if we could reuse this schedule for
the arm back-end as well
----------------------------------------------------------------
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]
