KellenSunderland commented on issue #16173: Saving and loading cudNN autotune and graph optimization URL: https://github.com/apache/incubator-mxnet/issues/16173#issuecomment-537796546 I agree this is an issue and really like the approach here @QueensGambit. TensorRT and Autotuning are the operations I see that impact our current cold start time. cuDNN is relatively straightforward to cache, TRT we might want to think about how to save and load the engine properly. I think the keys for caching are well though through. We would have to make sure the expectation is this will only work on a certain host-type, but I think that's a reasonable restriction.
---------------------------------------------------------------- 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: us...@infra.apache.org With regards, Apache Git Services