masahi commented on code in PR #15679:
URL: https://github.com/apache/tvm/pull/15679#discussion_r1324919768
##########
python/tvm/relax/transform/legalize_ops/manipulate.py:
##########
@@ -205,3 +213,16 @@ def te_layout_transform(data, name):
output_dtype = call_args[0].struct_info.dtype
output_sinfo = [TensorStructInfo(output_shape, output_dtype)]
return call_tir(gvar, call_args, output_sinfo, tir_vars)
+
+
+@register_legalize("relax.remove_pad")
+def _remove_pad(bb: BlockBuilder, call: Call) -> Expr:
+ orig_shape = call.attrs.orig_shape
+
+ def te_remove_pad(data):
+ """
+ Returns a new compute that restrict the original expression to the
shape of orig_shape
+ """
+ return te.compute(orig_shape, data, name="te_remove_pad")
Review Comment:
Better yet, is this actually needs to be an "op"? Since the op definition is
so trivial, how about moving this compute definition to `alter_op_layout.cc`
and use it there? You can remove all boilerplate and test cases required for a
Relax op definition.
--
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]
