yongwww commented on code in PR #18696:
URL: https://github.com/apache/tvm/pull/18696#discussion_r2742000849
##########
docs/how_to/tutorials/customize_opt.py:
##########
@@ -103,9 +109,17 @@ def forward(self, x):
# Import cublas pattern
-import tvm.relax.backend.cuda.cublas as _cublas
-
-
+try:
+ import tvm.relax.backend.cuda.cublas as _cublas
+except ImportError as e:
Review Comment:
probably we don't have to handle the tvm_ffi specifically here
##########
docs/how_to/tutorials/customize_opt.py:
##########
@@ -1,4 +1,4 @@
-# Licensed to the Apache Software Foundation (ASF) under one
+# Licensed to the Apache Software Foundation (ASF) under one
Review Comment:
pls take a look
##########
LICENSE:
##########
@@ -1,6 +1,6 @@
Apache License
Version 2.0, January 2004
- http://www.apache.org/licenses/
+ https://www.apache.org/licenses/
Review Comment:
The LICENSE update isn’t related to this PR. I’d suggest we avoid changing
it here and handle it in a separate PR if needed.
##########
docs/how_to/tutorials/customize_opt.py:
##########
@@ -55,6 +55,12 @@
from tvm import IRModule, relax
from tvm.relax.frontend import nn
+# Note: This tutorial requires TVM to be built with CUDA support.
+# If you encounter 'No module named ''tvm_ffi''' error,
Review Comment:
tvm_ffi can be installed with `cd 3rdparty/tvm-ffi; pip install .`
--
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]
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
