14 Commits

Author	SHA1	Message	Date
VictorSanh	290633b882	Fix args.gradient_accumulation_steps used before assigment.	2018-11-04 17:31:50 -05:00
VictorSanh	649e9774cd	Fix bug train_batch_size not an int. ... Division makes args.train_batch_size becoming a float. cc @thomwolf	2018-11-04 17:19:40 -05:00
VictorSanh	d55c3ae83f	Small logger bug (multi-gpu, distribution) in training	2018-11-04 16:28:10 -05:00
thomwolf	3ddff783c1	clean up + mask is long	2018-11-04 21:26:44 +01:00
thomwolf	c6207d85b6	remove old methods	2018-11-04 15:34:00 +01:00
thomwolf	965b2565a0	add distributed training	2018-11-04 15:32:04 +01:00
thomwolf	1ceac85e23	add gradient accumulation	2018-11-04 15:26:14 +01:00
thomwolf	6b0da96b4b	clean up	2018-11-04 15:17:55 +01:00
thomwolf	834b485b2e	logging + update copyright	2018-11-04 12:07:38 +01:00
thomwolf	1701291ef9	multi-gpu cleanup	2018-11-04 11:54:57 +01:00
thomwolf	f8276008df	update readme, file names, removing TF code, moving tests	2018-11-03 23:35:14 +01:00
thomwolf	4df6027453	clean up	2018-11-03 11:33:10 +01:00
Tim Rault	8163baab64	Convert indentation from 2 spaces to 4 spaces	2018-11-01 09:39:04 +01:00
thomwolf	13ee61e4de	switch to full google code	2018-10-31 18:46:03 +01:00