fastNLP

25 MB

Tree: e361b32c3a

Author	SHA1	Message	Date
yh_cc	e361b32c3a	在@linzehui 的帮助下seq2seq终于有了第一个版本; 目前实现了Seq2Seq的Transformer和LSTM版本，但metric和loss还没update;	5 years ago
yh_cc	ab047a2aa3	新增对RobertaEmbedding和GPT2Embedding的支持	5 years ago
yh_cc	7398f2b0b2	1.增加RobertaEmbedding与GPT2Embedding	5 years ago
yh_cc	45139ebbff	BertEmbedding支持选择是否缩小词表	5 years ago
yh_cc	ace8aafdb2	修复BertEmbedding在include_cls_sep=True时，SEP位置的值不正确的问题	5 years ago
yh_cc	367edb7ad1	修复move_to_device中device为None时可能出现错误的问题; BERTEmbedding中增加一个可选参数	5 years ago
yh_cc	8acb0f4ad6	1.新增Vocabulary的save与load功能; 2.修复BERTEmbedding drop_word的bug; 3.新增各种Embedding的使用注释	5 years ago
yh	5ebce3176f	增加部分测试	6 years ago
Yige Xu	880e3ad969	1. add mini_elmo.pkl and test codes for testing ElmoEmbedding; 2. update bert testing codes	6 years ago
yh_cc	e903db0e70	增加中文分类Pipe;使用矩阵加速BertEmbedding部分pool_method;调整部分测试用例名称;修复metric中对warning的误报	6 years ago
Yige Xu	4440801dbf	1. update bert.py and fix a bug in bert_embedding to adapt torch 1.2.0; 2. update models/bert.py and add BertForSentenceMatching model, now a BertEmbedding param should be passed to these five models; 3. create a small bert version for testing and modify test/models/test_bert.py; 4. move small glove and word2vec files to data_for_tests/embedding/small_static_embedding dir and fix relevant test codes; 5. delete some __init__.py files in test dir.	6 years ago
yh_cc	be77533c38	修复word drop bug, 增加相应测试	6 years ago
yh_cc	85f01f01d1	1.修复部分测试; 2.修复StaticEmbedding中未找到词初始化bug	6 years ago
yh	511f41dda1	1. 增加中文NER相关的loader和pipe; 2. 对应修改sequence_labeling的代码; 3.增加部分测试代码	6 years ago

14 Commits (e361b32c3a70b24bb9fa59e89a49039ce685a043)