Multi-thread training


I have been through the docs and it seems to be a command line interface option to use more than a single thread when training a model, docs here. But the corresponding rasa train --help does not display such option, I am using rasa 1.10.1 and Python 3.7.2. Is it a bug or am I doing something very wrongly?

Thanks for your help.

Hi @dan1,

while the reading of the training data and also the training of some policies isn’t multithreaded, all components / policies using tensorflow should automatically use multiple threads / CPUs.

Hi @Tobias_Wochinger,

Ok, great! Thanks for your help.