69 Commits

Author	SHA1	Message	Date
thomwolf	ae88eb88a4	set encoding to 'utf-8' in calls to open	2018-12-14 13:48:58 +01:00
thomwolf	52c53f39d0	clean up apex integration	2018-12-13 13:02:17 +01:00
thomwolf	d23eed85bb	model loading apex modification	2018-12-13 12:53:17 +01:00
thomwolf	93f335ef86	add pretrained loading from state_dict	2018-12-13 12:48:13 +01:00
Thomas Wolf	91aab2a6d3	Merge pull request #116 from FDecaYed/deyuf/fp16_with_apex ... Change to use apex for better fp16 and multi-gpu support	2018-12-13 12:32:37 +01:00
Deyu Fu	3b0a14b761	add fallback path for apex used in modeling.py	2018-12-12 15:05:45 -08:00
Deyu Fu	c8ea286048	change to apex for better fp16 and multi-gpu support	2018-12-11 17:13:58 -08:00
thomwolf	270fa2f20b	add pretrained loading from state_dict	2018-12-11 11:50:38 +01:00
Grégory Châtel	fc5a38ac92	Adding the BertForMultipleChoiceClass.	2018-12-06 18:42:23 +01:00
thomwolf	511bce58bd	update new token classification model	2018-11-30 22:56:02 +01:00
thomwolf	d787c6be8c	improve docstrings and fix new token classification model	2018-11-30 22:55:26 +01:00
thomwolf	d6f06c03f4	fixed loading pre-trained tokenizer from directory	2018-11-30 14:09:06 +01:00
thomwolf	532a81d3d6	fixed doc_strings	2018-11-30 13:57:01 +01:00
thomwolf	296f006132	added BertForTokenClassification model	2018-11-30 13:56:53 +01:00
thomwolf	298107fed7	Added new bert models	2018-11-30 13:56:02 +01:00
thomwolf	63ae5d2134	added cache_dir option in from_pretrained	2018-11-26 10:21:56 +01:00
thomwolf	ebaacba38b	fixing typo in docstring	2018-11-26 09:55:15 +01:00
thomwolf	870d71636e	fixing target size in crossentropy losses	2018-11-26 09:51:34 +01:00
thomwolf	1de35b624b	preparing for first release	2018-11-15 20:56:10 +01:00