-
Notifications
You must be signed in to change notification settings - Fork 3.5k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[CI] Fix gluoncv tutorial under mxnet-mkl #5458
Comments
cc @icemelon9 @shoubhik @anijain2305 |
@icemelon9 Do you know about this operator? |
It's weird that I cannot reproduce the error in my local environment with mxnet-mkl 1.6.0 and gluoncv. |
@icemelon9 could it due to that mxnet optionally enabled amp on the platform that support these ops? |
@tqchen Yes, that might be possible. I'll check it on a GPU instance. |
Ping @icemelon9 If you get time. |
I cannot reproduce the error either on Ubuntu GPU instance
for both mxnet==1.6.0 and mxnet-mkl==1.6.0 |
This is an error message generated by the mxnet-mkl image in the current master. The error was due to the operator
amp_multicast
is not implmented in the mxnet frontend(likely due to a mkl specialization).http://ci.tvm.ai:8080/job/temp-ci-docker-staging/job/ci-stage2/6/execution/node/309/log/?consoleFull
Please see if we are interested in quick a fix, since we have docker upgrades depending on it. In the meanwhile, I will revert the docker image to mxnet for now until the next upgrade.
The text was updated successfully, but these errors were encountered: