Reproducibility is important in a production ML setting. Majority of the companies don't have necessary tools nor they see the value in it. Also limitations like nondeterminism in CUDA kernels doesn't help. The recent trend in containerizing ML code greatly helps reproducibility.
#data isn't much without evaluation platforms http://mlperf.org @agramfort at reproducibilty workshop #ICML2018
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.