hummingbird
/
fastNLP
Not watched
2
0
Fork 0
Code
Releases 13 Wiki Activity
Issues 0 Pull Requests 0 Datasets Model Cloudbrain

Commit Graph

  • *
  • *
  • |\
  • * |
  • | | *
  • | | | *
  • | | | *
  • | | | *
  • | |_|/
  • |/| |
  • * | |
  • |\ \ \
  • | * | |
  • | | | *
  • | | | *
  • | | | |\
  • | | | * |
  • | | | | *
  • * | | | |
  • |\ \ \ \ \
  • | * | | | |
  • |/ / / / /
  • | * / / /
  • |/ / / /
  • | | | *
  • | | | *
  • | | |/
  • | | *
  • | | *
  • | | *
  • | | *
  • * | |
  • |\ \ \
  • | * | |
  • | | | *
  • | | | *
  • | | | *
  • | | | *
  • | | |/
  • | * /
  • |/ /
  • * |
  • |\ \
  • | | *
  • | | |\
  • | | * |
  • | | | *
  • | | |/
  • | * /
  • |/ /
  • | *
  • | *
  • | *
  • | |\
  • | * |
  • * | |
  • |\ \ \
  • | | |/
  • | |/|
  • | * |
  • | * |
  • | * |
  • | * |
  • | |/
  • | *
  • | *
  • | *
  • | *
  • | *
  • | *
  • | |\
  • | | *
  • | * |
  • | |\ \
  • | | * |
  • | | * |
  • | | | *
  • | | * |
  • | | * |
  • | | | *
  • | | | *
  • | | |/
  • | |/|
  • | * |
  • | |/
  • | *
  • | *
  • | *
  • | *
  • | *
  • | *
  • | *
  • | |\
  • | * |
  • | * |
  • | | *
  • | |/
  • | *
  • | |\
  • | * |
  • | | *
  • | | *
  • | | *
  • | | *
  • | |/
  • * |
  • * |
  • |\|
  • | *
  • | *
  • | *
  • | *
  • | |\
  • | * |
  • | | *
  • | | *
  • | | |\
  • | |_|/
  • |/| |
  • | | *
  • | | *
  • | | *
  • | | |\
  • | | * |
  • | | | *
  • | | | |\
  • | | | |/
  • | | |/|
  • | | * |
  • | | | *
  • | | |/
  • | | *
  • | | |\
  • | | * |
  • | |/ /
  • | | *
  • | | |\
  • | | |/
  • | |/|
  • | | *
  • * | |
  • |\ \ \
  • | | * |
  • | | |/
  • | | *
  • | | |\
  • | | * |
  • | | * |
  • 58f373d3 - fix test by yunfan 2019-04-12 15:55:22 +0800
  • 6d11ef86 Merge remote-tracking branch 'private/dev' into pr by yunfan 2019-04-12 15:40:43 +0800
  • 70fb4a22 - add star transformer model - add ConllLoader, for all kinds of conll-format files - add JsonLoader, for json-format files - add SSTLoader, for SST-2 & SST-5 - change Callback interface - fix batch multi-process when killed - add README to list models and their performance by yunfan 2019-04-12 15:35:22 +0800
  • 4fd49cc3 add sigmoid activate function in MLP by xuyige 2019-04-11 15:00:10 +0800
  • 0f8bed73 Update README.md by Xipeng Qiu 2019-04-08 20:14:46 +0800
  • c344f7de Update README.md by Xipeng Qiu 2019-04-08 20:14:12 +0800
  • 667b3126 Update README.md by Xipeng Qiu 2019-04-08 20:11:47 +0800
  • 90d112c0 Merge pull request #141 from wlhgtc/master by Yunfan Shao 2019-03-31 13:20:20 +0800
  • 28d3f500 Fix bug in MSA2 (mixed k and v) by wlhgtc 2019-03-31 00:00:59 +0800
  • 55f65c39 add SQuAD metric by xuyige 2019-03-27 17:04:19 +0800
  • ec90a1f0 Merge branch 'dev' of github.com:choosewhatulike/fastNLP-private into dev by yh 2019-03-23 18:13:35 +0800
  • e5f68bbd 修复CRF为负数的bug by yh 2019-03-23 18:12:32 +0800
  • 6a498bbd * 给vocabulary添加遍历方法 by FengZiYjun 2019-03-23 15:44:23 +0800
  • 56410c9e Merge pull request #139 from fastnlp/choosewhatulike-patch-1 by Yunfan Shao 2019-03-23 13:31:04 +0800
  • b7008cba fix mask bug in star-transformer by Yunfan Shao 2019-03-22 17:37:50 +0800
  • 8d61cd68 fix the "masked_fill" bug by wlhgtc 2019-03-22 17:05:43 +0800
  • f5ab7a5d * 将enas相关代码放到automl目录下 * 修复fast_param_mapping的一个bug * Trainer添加自动创建save目录 * Vocabulary的打印，显示内容 by FengZiYjun 2019-03-21 15:03:51 +0800
  • ef0c6e93 Changes to Callbacks: * 给callback添加给定几个只读属性 * 通过manager设置这些属性 * 代码优化，减轻@transfer的负担 by FengZiYjun 2019-03-20 09:49:01 +0800
  • 99d6bb20 change two default init arguments of Trainer into None by FengZiYjun 2019-03-16 19:54:03 +0800
  • f2d7d01b 修复CRF中可能存在的bug by yh 2019-03-13 10:16:39 +0800
  • 22661ea8 修复crf中typo; 以及可能导致数值不稳定的地方 by yh 2019-03-12 19:51:28 +0800
  • 3d7cfb35 修复fieldarray中padder潜在bug by yh 2019-03-12 19:00:50 +0800
  • 88d4de7c Merge pull request #135 from choosewhatulike/pr by Xipeng Qiu 2019-03-11 21:29:58 +0800
  • 7c7f28f2 - add star-transformer by yunfan 2019-03-10 21:53:44 +0800
  • 95a72f06 * AutoPadder will not pad when dtype is None * add ignore_type in DataSet.apply by FengZiYjun 2019-02-25 15:38:38 +0800
  • 3e01b142 add ignore_type in DataSet.add_field by FengZiYjun 2019-02-25 10:26:03 +0800
  • 251088ac Add ENAS (Efficient Neural Architecture Search) by chenkaiyu1997 2019-02-15 17:31:56 +0800
  • feb8f63a Update README.md by Coet 2019-01-14 09:31:35 +0800
  • 5241e30b - update transformer docs by yunfan 2019-02-20 17:27:36 +0800
  • 767e7971 Merge pull request #134 from chenkaiyu1997/master by Coet 2019-02-18 09:59:14 +0800
  • f14d3175 Merge branch 'dev' of github.com:choosewhatulike/fastNLP-private into dev by yh 2019-02-17 14:16:33 +0800
  • 8d4f26bb 增加metric注释；修改trainer save过程中的bug by yh 2019-02-17 14:16:19 +0800
  • ee677d5d update MLP module by xuyige 2019-02-17 02:12:33 +0800
  • efeac2c4 Add ENAS (Efficient Neural Architecture Search) by chenkaiyu1997 2019-02-15 17:31:56 +0800
  • 3ea7de16 1.修改ClipGradientCallback的bug；删除LRSchedulerCallback中的print，之后应该传入pbar进行打印;2.增加MLP注释 by yh 2019-02-14 13:18:50 +0800
  • 5eb126db BucketSampler增加一条错误检测 by yh 2019-02-09 13:47:13 +0800
  • 853bea58 Merge branch 'dev' of github.com:choosewhatulike/fastNLP-private into dev by yh 2019-02-06 22:26:19 +0800
  • 29eab18b 1. CRF增加支持bmeso类型的tag 2. vocabulary中增加注释 by yh 2019-02-06 22:26:10 +0800
  • 13faa2b4 (tag: v0.3.1) Merge pull request #132 from FengZiYjun/v0.3.1 by Coet 2019-02-06 10:40:53 +0800
  • b66d7b8f update API introduction by FengZiYjun 2019-02-04 10:07:56 +0800
  • d1b5adab add codecov fix by FengZiYjun 2019-02-04 09:56:08 +0800
  • 0c5630bd Ready for V0.3.1 * 升级parser API和模型 * update docs: add new pages for tutorials * upgrade CWS api download source * add a new method for dataset field access * add introduction for bert * add more unit tests for api/processor * remove unused test data. Add new test data. by FengZiYjun 2019-02-04 09:44:54 +0800
  • 98654113 整理所有dataset loader，建立单元测试 by FengZiYjun 2019-02-02 16:46:42 +0800
  • bfaf09df add BERT model * load pre-trained BERT weights from local binary * add tests by FengZiYjun 2019-01-29 20:35:12 +0800
  • 887fc928 update callbacks: * rename callback methods. Use fastai's notation. * add a new callback method - on_valid_begin by FengZiYjun 2019-01-25 21:43:24 +0800
  • e0d6a259 skip training while n_epoch in trainer is not greater than 0 by xuyige 2019-01-23 17:09:28 +0800
  • d4b4ffa2 add testing tutorial by xuyige 2019-01-23 14:56:25 +0800
  • 9474ab4b remove device in batch by yunfan 2019-01-21 22:28:31 +0800
  • c02980e0 Merge branch 'yyff' into dev by yunfan 2019-01-21 14:55:53 +0800
  • a37de434 add batch device by yunfan 2019-01-21 14:50:41 +0800
  • e93c6f00 Merge branch 'dev' of https://github.com/choosewhatulike/fastNLP-private into dev by yunfan 2019-01-21 14:52:24 +0800
  • 47ec69ea trainer根据syf的多进程batch进行修改 by yh 2019-01-21 14:44:31 +0800
  • f3cb8125 将tesorboardX处理为callback, 从trainer移除tensorboardX相关代码 by FengZiYjun 2019-01-20 16:37:58 +0800
  • 03f49c82 - batch with multiprocessing by yunfan 2019-01-19 19:44:32 +0800
  • b14dd588 Update POS API by FengZiYjun 2019-01-19 18:48:57 +0800
  • 62ea4f7f 添加LR finder，用第一个epoch找最佳lr,从第二个epoch开始训练 by FengZiYjun 2019-01-19 18:40:43 +0800
  • a7f3701b - revert batch by yunfan 2019-01-19 16:26:39 +0800
  • de856fb8 update reproduction by yunfan 2019-01-15 15:33:39 +0800
  • eb55856c - fix parser train by yunfan 2019-01-19 16:07:10 +0800
  • ab953b43 * 重构POS API，改成接受word作为输入 * 添加两类Callback * 完善Trainer对error的捕捉 by FengZiYjun 2019-01-19 15:23:07 +0800
  • d9ac3344 减少batch中不断创建多进程的开销 by yh_cc 2019-01-18 23:33:19 +0800
  • 2e3ef52a 将batch增强为多进程batch by yh_cc 2019-01-18 23:02:15 +0800
  • 864c2238 添加FieldArray对list of np.array的支持 by FengZiYjun 2019-01-17 22:42:40 +0800
  • b93ca9bb * FieldArray添加对list of np.array的支持 * 添加测试：FieldArray的初始化 by FengZiYjun 2019-01-17 15:39:13 +0800
  • e4f997d5 refactor type system in FieldArray: * 重构dtype的检测代码，在FieldArray的初始化和append两处，达到更好的代码复用 * 类型检测的责任完全落在FieldArray，DataSet与之配合 测试： * 整理dtype相关的测试代码 * 给所有tutorial添加测试 其他： * 完善一个完整的Conll dataset loader * 升级POS tag model训练脚本 by FengZiYjun 2019-01-17 12:25:37 +0800
  • 73dd35da Merge branch 'dev' of github.com:choosewhatulike/fastNLP-private into dev by yh 2019-01-15 22:23:35 +0800
  • 3e33a230 修改Padder的测试用例 by yh 2019-01-15 22:23:19 +0800
  • 8091a734 1. 将pad的功能从FieldArray中剥离出来，使用Padder完成各种padding操作。 2. FieldArray默认使用AutoPadder, AutoPadder的行为与之前不使用padder是一致的的 3. 为了解决二维padding的问题，引入了EngChar2dPadder用于对character进行padding 4. 增加一份padding的tutorial。 by yh 2019-01-15 22:21:55 +0800
  • d80d944e * 添加callbacks：EarlyStopCallback * 将dataset.py中的assert改为raise error * 给trainer添加try-except,捕捉EarlyStopError * 优化trainer代码 * 给callbacks添加测试 by FengZiYjun 2019-01-15 17:08:53 +0800
  • 1f50b01f conflict solved by yh 2019-01-15 15:16:20 +0800
  • 6a0a1ed4 train增加注释；attention增加注释；新增transformer分词 by yh 2019-01-15 14:58:43 +0800
  • 1fdaf236 Updates: * 改名: chinese_word_segment ---> Chinese_word_segmentation * 改名: pos_tag_model ---> POS_tagging * 添加4个对Batch的测试 * 删除无用的chinese_word_segment/run.py by FengZiYjun 2019-01-15 14:56:01 +0800
  • c4ba75d1 code optimization * move used readers from reproduction to io/dataset_loader.py (API shall not call anything from reproduction/) by FengZiYjun 2019-01-15 14:30:37 +0800
  • a6dbbe98 remove the gpu_id info when saving by ChenXin 2019-01-15 11:45:02 +0800
  • 2e9e6c6c - fix trainer with validate_every > 0 - refine & fix Transformer Encoder - refine & speed up biaffine parser by yunfan 2019-01-14 19:13:52 +0800
  • 3fa95b6c Update README.md by Coet 2019-01-14 09:31:35 +0800
  • 3219b9da (tag: v0.3.0) Merge pull request #123 from FengZiYjun/v0.3.0 by Coet 2019-01-14 09:07:31 +0800
  • 094a5661 add syntactic parsing API by FengZiYjun 2019-01-13 11:49:38 +0800
  • 8df5bce9 fastNLP V0.3 by FengZiYjun 2019-01-12 19:15:20 +0800
  • eca0c6f8 Prepare for release. * pass all tests * prepare CWS & POS API * update tutorials * add README.md in tutorials/ & api/ by FengZiYjun 2019-01-12 19:00:06 +0800
  • 62a7556a Merge remote-tracking branch 'private/dev' into dev by yunfan 2019-01-12 11:26:32 +0800
  • ba28702e update Biaffine Parser, Variational RNN add parser API by yunfan 2019-01-12 11:22:09 +0800
  • 7eb02f17 update code comments in CRF by FengZiYjun 2019-01-11 20:59:42 +0800
  • 751fe276 Merge branch 'master' of https://github.com/fastnlp/fastNLP into current branch by FengZiYjun 2019-01-11 20:44:26 +0800
  • c0b67a2b fix tests by FengZiYjun 2019-01-11 20:24:35 +0800
  • 806364b5 * update travis config * update README by FengZiYjun 2019-01-11 20:14:18 +0800
  • a1fe24c4 Merge branch 'dev' of github.com:choosewhatulike/fastNLP-private into dev by yh 2019-01-11 19:55:54 +0800
  • 145125fe 支持CWS的高级api“ by yh 2019-01-11 19:55:38 +0800
  • c8e6497f * 调整测试代码，通过全部测试 * 合并callback和trainer的更新 * CWS API的更新 by FengZiYjun 2019-01-11 19:47:07 +0800
  • 40055297 1. callback中增加GradientClip; 2.Trainer中取消_print_train()和_tqdm_train()，全部并入了_train() by yh 2019-01-11 19:09:01 +0800
  • e8ea6ea3 improved callbacks & trainer code structure by FengZiYjun 2019-01-11 18:14:58 +0800
  • 59c9b894 Merge branch 'dev' of https://github.com/choosewhatulike/fastNLP-private into callbacks by FengZiYjun 2019-01-11 14:43:54 +0800
  • a7274c66 update callbacks in Trainer by FengZiYjun 2019-01-11 14:43:06 +0800
  • 12356b4c Merge branch 'dev' of https://github.com/choosewhatulike/fastNLP-private into dev by xuyige 2019-01-10 18:05:11 +0800
  • 409e25e1 update fastnlp_advanced_tutorial by xuyige 2019-01-10 18:04:53 +0800
  • ef82c1f1 Merge pull request #121 from FengZiYjun/doc-fixing by Coet 2019-01-10 12:50:46 +0800
  • 179d1232 add callback in Trainer by FengZiYjun 2019-01-09 22:42:07 +0800
  • 8f1947d1 Merge remote-tracking branch 'origin/dev' into dev by xuyige 2019-01-09 00:11:39 +0800
  • d978b0c1 update ESIM model and rename it by xuyige 2019-01-09 00:11:05 +0800
  • a97bdc7f print type while printing instance by xuyige 2019-01-09 00:10:30 +0800