Tianqi Chen
Tianqi Chen
I agree that evaluate more aspects. Some of them is already covered in this proposal, for example - Memory consumption - Parallelization and Scheduling overhead One of the most important...
echo @bhack I have some thoughts in turns of standardize the frameworks. It is fundamentally hard for all frameworks to be forced into one. However, we might have enough idea...
This was due to incorrect timing on async API... there is no magic to run much faster than others using CuDNN..
Would be good to get a status update @junrushao1994 . I would suggest we move followup non-infra part to separate trackings to keep things tracable.
Would be good to know why. To see if it is an issue with particular thread pool, we can try to switch to openmp https://github.com/apache/tvm/blob/main/cmake/config.cmake#L178
That might hint something is wrong higher up?
LGTM! @kparzysz-quic do you mind also post an announcement thread? So people are aware
Alternatively, we can make name supply an `idempotent` object, which means it is not part of the serialization and will be nullptr initially. Calling an API `IRModule->GetOrCreateNameSupply()` will recreate it...
Thanks @gigiblender Adding module name to IRModule as an optional attr sounds great. BTW, we still need to distinguish the case where the user expect "the name". This is different...
We should perhaps close this tracking issue?