test/fastNLP
1
0
Fork 0
mirror of https://gitee.com/fastnlp/fastNLP.git synced 2024-12-02 20:27:35 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
1,298 Commits 3 Branches 13 Tags 25 MiB
1756e3ffdf
Commit Graph

1298 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
yh_cc
1756e3ffdf 1.修复MNLILoader中的bug; 2.修复field中的tensor warning 2019-08-29 11:16:59 +08:00
yh_cc
146a004dee 修改travis converage设置 2019-08-29 10:12:30 +08:00
yhcc
09d0b74595
Update .travis.yml 
TRAVIS默认已经加入了
 2019-08-29 09:56:36 +08:00
Yige Xu
39de27f472 Update BertModel.from_pretrained function. Now can pass a model_dir_or_name instead of model_dir. 2019-08-29 08:19:36 +08:00
xuyige
8142bad87a Merge branch 'dev0.5.0' of https://github.com/fastnlp/fastNLP into dev0.5.0 2019-08-29 01:43:19 +08:00
xuyige
5d8a8c98c6 1. delete io/data_loader dir; 2. delete model/enas*; 3. delete legacy dir; 4. delete DateSetLoader and relevant codes; 5. fix a test code error in core/test_dataset.py; 6. delete io.BaseLoader and relevant code. 2019-08-29 01:43:04 +08:00
yh_cc
cbe5b347e5 SpanFMetric增加对encoding_type和tag_vocab的检查 2019-08-28 23:53:53 +08:00
yh_cc
55e736bf4c SpanFMetric增加对encoding_type和tag_vocab的检查 2019-08-28 23:53:20 +08:00
yh_cc
a46b8f129b Trainer中保存最佳模型存在bug 2019-08-28 23:06:13 +08:00
yh_cc
6201f66178 Trainer中保存最佳模型存在bug 2019-08-28 22:56:02 +08:00
ChenXin
fbbb2fcd8e fix some bugs in docs 2019-08-27 21:33:18 +08:00
ChenXin
169f519ffb ignore the methods inherited from torch.nn.Embedding 2019-08-27 21:07:22 +08:00
ChenXin
04737a105d update the doc-tool to show __init__ and class doc separately 2019-08-27 20:46:05 +08:00
yh
359a176748 merge conflict 2019-08-27 01:57:33 +08:00
yh
19bbaf11b6 使用更pytorch的方式处理embedding中的parameter 2019-08-27 01:54:15 +08:00
ChenXin
53975c045a update the doc-tool & fix an importing bug 2019-08-26 14:58:36 +08:00
zide05
78af3491a4 修改tutorial 2019-08-26 14:03:40 +08:00
ChenXin
ffd5fd8135 delete the old doc-tool 2019-08-26 11:58:20 +08:00
ChenXin
e1f234841c mark the dataloader.__init__ as undocumented 2019-08-26 11:00:45 +08:00
ChenXin
2cf9c0ebb1 add __all__ and __doc__ for all files in module 'modules', using 'undocumented' tags 2019-08-26 10:26:55 +08:00
ChenXin
efa9496d09 add __all__ and __doc__ for all files in module 'models', using 'undocumented' tags 2019-08-26 10:25:51 +08:00
ChenXin
0d5f43b451 add __all__ and __doc__ for all files in module 'io', using 'undocumented' tags 2019-08-26 10:25:01 +08:00
ChenXin
efe88263bb add __all__ and __doc__ for all files in module 'core', using 'undocumented' tags 2019-08-26 10:21:10 +08:00
ChenXin
9535ec60b6 update the doc system: use customized tool to generate the rst files 2019-08-26 10:17:43 +08:00
ChenXin
34e17e9793 update the fastNLP.__init__ : use loader&pipe to replace data_loader 2019-08-26 10:07:52 +08:00
ChenXin
9e16791c53 fix some importing bugs 2019-08-26 08:32:39 +08:00
xuyige
78be840ab9 1.update README 2. fix a filename-bug in pretrain_static_file; 3. add Pipe to documents; 4. update documents in some loaders; 5. update tutorial 2 & 3 to adapt version 0.5.0 2019-08-26 01:56:20 +08:00
yh_cc
584a92c64c 1.增加sequence labeling任务的数据说明; 2.增加对CWSPipe的引用 2019-08-26 01:33:17 +08:00
yh_cc
be77533c38 修复word drop bug, 增加相应测试 2019-08-25 18:58:03 +08:00
yh_cc
74934271dc 1.增加sequence labelling中bert ner; 2.将print替换为logger 2019-08-25 18:20:58 +08:00
xuyige
65a6fd3dc7 Revert "delete predictor.py" 
This reverts commit 8445bdbc79.
 2019-08-25 17:48:01 +08:00
ChenXin
8445bdbc79 delete predictor.py 2019-08-25 16:57:47 +08:00
yhcc
f3a2aa6da5
Merge pull request #208 from SrWYG/dev0.5.0 
[verify] char_cnn use pipe
 2019-08-25 15:55:16 +08:00
ChenXin
d6c597d32e add __doc__ & __all__ in module 'embeddings' 2019-08-25 11:13:25 +08:00
yh
9f3336f167 Merge branch 'dev0.5.0' of https://github.com/fastnlp/fastNLP into dev0.5.0 2019-08-24 01:09:40 +08:00
yh
8d7c3ba140 修复CharacterEmbedding中的bug 2019-08-24 01:09:28 +08:00
wyg
ed6fd60aa9 [verify] char_cnn use pipe 2019-08-23 14:47:46 +08:00
yh_cc
85f01f01d1 1.修复部分测试; 2.修复StaticEmbedding中未找到词初始化bug 2019-08-23 11:08:28 +08:00
yh_cc
c38e8986cc 在linux桌面系统上Trainer中使用Tester的tqdm存在bug; 增加一个可选项使得用户可以关闭Tester的tqdm 2019-08-22 19:20:24 +08:00
yh_cc
f18ab642d7 pytorch1.2版本中新增boolTensor类型，所有的masked_fill必须为ByteTensor类型的索引,修改fastNLP以适配 2019-08-22 15:51:44 +08:00
yh
e2232ac39f 1.修复Embedding中潜在的寻找路径失败; 2.reproduction中增加分词模型; 3.修改ner模型为最新的pipe版本 2019-08-20 21:35:12 +08:00
yunfan
44d569dade [fix] logger in dist_trainer 2019-08-20 16:28:12 +08:00
yh
a81c839fd3 Merge branch 'dev0.5.0' of https://github.com/fastnlp/fastNLP into dev0.5.0 2019-08-20 16:04:54 +08:00
yh
ce083de26b 1.删除Trainer中的prefetch参数; 2.增加中文分词的下载; 3.增加DataBundle的delete_dataset, delete_vocab 2019-08-20 16:04:51 +08:00
yh
7a0903d9ba 1.删除Trainer中的prefetch参数; 2.增加中文分词的下载; 3.增加DataBundle的delete_dataset, delete_vocab 2019-08-20 16:04:28 +08:00
yunfan
f7ebc1ca2c Merge branch 'dev0.5.0' of https://github.com/fastnlp/fastNLP into pr 2019-08-20 15:28:35 +08:00
yunfan
4e59d88724 [update] move logger to fastNLP.core, update logging format 2019-08-20 15:28:21 +08:00
yh
f69b50c8f3 Merge branch 'dev0.5.0' of https://github.com/fastnlp/fastNLP into dev0.5.0 2019-08-20 15:06:32 +08:00
yh
32a2f197e1 增加对metric name设置的支持 2019-08-20 15:06:01 +08:00
ChenXin
d0354d8e28 fix some importing bugs 2019-08-20 00:00:31 +08:00