sandeep-krishnamurthy commented on a change in pull request #12691:
[MXNET-1001] Add new cuddn flag in Gluon Conv blocks to control - cudnn_tune,
cudnn_off params
URL: https://github.com/apache/incubator-mxnet/pull/12691#discussion_r234046370
##########
File path: tests/python/gpu/test_gluon_gpu.py
##########
@@ -408,6 +431,60 @@ def tensor_size(big_tensor_bytes):
# Evaluate model
net(data_in).asnumpy()
+@with_seed()
+def test_conv_invalid_cudnn_flag():
+ try:
+ _check_layer_forward(nn.Conv1D(16, 3, in_channels=4, cudnn='fail_me'),
(1, 4, 10))
+ assert False, "Gluon Conv1D block did not raise the AssertionError for
invalid cudnn flag."
+ except AssertionError:
Review comment:
Thanks. updated.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
