Lunderberg commented on issue #8690: URL: https://github.com/apache/tvm/issues/8690#issuecomment-895322905
Hmm, that failing test is parametrized to include `float16`, but the failing case was using `float32` values. It's also running with a fixed random seed to try to prevent any flakiness due to input data. That particular parametrization was running with cudnn, added to the CI in #8383 and recently merged. I'm not able to reproduce the failure in my local `ci_gpu` docker instance, either before or after #8383. (Ran 100 iterations just no, no issues locally.) -- 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]
