olojuwin opened a new issue #16485: cudnn_convolution-inl.h:160: Check failed: e == CUDNN_STATUS_SUCCESS (8 vs. 0) : cuDNN: CUDNN_STATUS_EXECUTION_FAILED URL: https://github.com/apache/incubator-mxnet/issues/16485 work environment：2080TI CUDA:9.0 MXNET_VERSION:1.2.0 ,1.2.1 1.5.0 CUDNN: NO when I train resnet101 ,if i set export MXNET_CPU_WORKER_NTHREADS=24, I'll get this mistake. If I don't set MXNET_CPU_WORKER_NTHREADS，it work good. If I change it to something else backbone ，it work good. if i change CUDA to 10.0,it work good.
---------------------------------------------------------------- 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. For queries about this service, please contact Infrastructure at: [email protected] With regards, Apache Git Services
