adstraw commented on a change in pull request #9473:
URL: https://github.com/apache/tvm/pull/9473#discussion_r746099503
##########
File path: tests/python/contrib/test_hexagon/infrastructure.py
##########
@@ -18,36 +18,43 @@
""" Hexagon testing infrastructure """
import tvm
+from tvm import te
import numpy
def ceildiv(o, d):
return tvm.tir.floordiv(o + d - 1, d)
-def get_packed_activation_layout(shape_nhwc, block_shape, packed_C=True):
+def get_block_shape():
+ return 8, 8, 32
+
+
+def get_filter_block_shape():
+ return 8, 32, 4
+
+
+def get_packed_shape(shape_nhwc):
Review comment:
Renamed to `logical_shape_nhwc`. Also renamed other shape parameters in
the infrastructure file for clarity.
In regards to padding the role of this function is to convert the
post-padded logical shape into a physical packed shape. I think the intent of
this function is clear, now, given the variable name change. Also added
comments to this effect.
##########
File path: tests/python/contrib/test_hexagon/infrastructure.py
##########
@@ -18,36 +18,43 @@
""" Hexagon testing infrastructure """
import tvm
+from tvm import te
import numpy
def ceildiv(o, d):
return tvm.tir.floordiv(o + d - 1, d)
-def get_packed_activation_layout(shape_nhwc, block_shape, packed_C=True):
+def get_block_shape():
+ return 8, 8, 32
+
+
+def get_filter_block_shape():
+ return 8, 32, 4
+
+
+def get_packed_shape(shape_nhwc):
Review comment:
Done.
--
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]
