3 Commits

Author	SHA1	Message	Date
Aapo Kyrola	3410939459	pass learning rate scaling factor to parameter update builder function ... Summary: When refactoring data parallel model, the division of LR by number of devices was dropped, and thus we ended up effectively multiplying gradients by the number of devices. Thus, we need to scale the LR by 1/numgpus. Created a test to confirm that data_parallel_model produces exactly same results on different number of gpus, given the total batch size. Reviewed By: prigoyal Differential Revision: D4248907 fbshipit-source-id: af21ede113e6ac25f12c556de298cb18974548be	2016-12-05 11:53:26 -08:00
Yangqing Jia	238ceab825	fbsync. TODO: check if build files need update.	2016-11-15 00:00:46 -08:00
Yangqing Jia	d1e9215184	fbsync	2016-10-07 13:08:53 -07:00