test/fastNLP
1
0
Fork 0
mirror of https://gitee.com/fastnlp/fastNLP.git synced 2024-12-02 12:17:35 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
153 Commits 3 Branches 13 Tags 25 MiB
ffe7c26369
Commit Graph

153 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
Coet
ffe7c26369
Merge branch 'master' into more_code_comments 2018-08-16 21:46:36 +08:00
Xipeng Qiu
abbe423dbe
Update README.md 2018-08-16 15:42:27 +08:00
choosewhatulike
1146ef0825 fix test_metrics 2018-08-15 21:41:10 +08:00
Coet
964cc2dc26
Merge pull request #25 from fastnlp/create_issue_templates 
Update issue templates
 2018-08-15 21:30:45 +08:00
FengZiYjun
929a595c4c Merge branch 'master' of https://github.com/fastNLP/fastNLP into to_merge 
# Conflicts:
#	fastNLP/core/metrics.py
#	fastNLP/core/predictor.py
 2018-08-15 21:07:55 +08:00
FengZiYjun
4bbeaebe96 Updates to cores, action, loader: 
- rename Inference to Predictor
- rename Trainer.prepare_input to Trainer.load_train_data, load data_train.pkl only
- add __contains__ method to config Section class
- more code comments
- more elegant make_batch & data_iterator: Samplers return batch samples instead of batch indices
 2018-08-15 20:12:20 +08:00
choosewhatulike
8a87807274 set no_grad() for test & inference, reduce memory usage 2018-08-14 00:35:48 +08:00
Yunfan Shao
762a559fab fix bug in SeqLabelTester 2018-08-14 00:14:13 +08:00
Coet
4bc123dc16 Update issue templates 2018-08-13 20:26:16 +08:00
choosewhatulike
d6ef132207 fix test_metrics bug 2018-08-12 20:04:38 +08:00
choosewhatulike
2961afd326 remove unused file 2018-08-12 19:55:19 +08:00
Coet
0b2e2ff586
Merge pull request #23 from fastnlp/dev/ci_test 
Dev/ci test
 2018-08-12 18:30:45 +08:00
Coet
8a6976460a
Merge pull request #24 from fastnlp/add-code-of-conduct-1 
Create CODE_OF_CONDUCT.md
 2018-08-12 18:16:00 +08:00
Coet
1cd7325328
Create CODE_OF_CONDUCT.md 2018-08-11 21:50:54 +08:00
Coet
3560fb1f67
Merge pull request #22 from fastnlp/dev/ner 
Dev/ner
 2018-08-11 21:29:24 +08:00
FengZiYjun
380fc251f4 add code coverage 2018-08-11 10:11:09 +08:00
FengZiYjun
77b3a0c67d fastNLP high-level interface: 
- fastNLP interface for sequence labeling works
- fastNLP interface for text classification works
 2018-08-11 09:20:09 +08:00
Coet
2075693273
Merge pull request #21 from fastnlp/dev/classify 
Dev/classify
 2018-08-10 19:45:37 +08:00
FengZiYjun
80a127cb24 merge jianghao's code 2018-08-09 21:57:53 +08:00
FengZiYjun
8e6db05339 changes to Trainer, Tester & Inference: 
- rename "POSTrainer", "POSTester" to "SeqLabelTrainer", "SeqLabelTester"
- Trainer & Tester have NO relation with Action
- Inference owns independent "make_batch" & "data_forward"
- Conversion to Tensor & go into cuda are done in "make_batch"
- "make_batch" support maximum/minimum length
 2018-08-08 20:40:44 +08:00
FengZiYjun
c1d7c5d7da changes to action, trainer and tester: 
- rename "POSTrainer" to "SeqLabelTrainer"
- add text classification test data
- update make_batch in  Trainer and Tester
 2018-08-07 19:18:56 +08:00
Yunfan Shao
cd5c1f41fb
Merge pull request #19 from choosewhatulike/master 
add metrics
 2018-08-06 22:00:10 +08:00
Yunfan Shao
defdfaadd2
Merge pull request #20 from choosewhatulike/fix_optim 
fix optimizer
 2018-08-06 21:58:57 +08:00
FengZiYjun
83f69b0e0f Combine make_batch for Trainer and Tester 
- change parameter <seq_length-->mask> in loss function defined in seq model
- Trainer & Tester have Action as default parameter, shared static methods like make_batch
- add seq_len in make_batch of Inference
- add SeqLabelInfer, a subclass of Inference
- seq_labeling.py works
 2018-08-06 20:15:31 +08:00
FengZiYjun
2c0079f3d5 changes to action, trainer and tester: 
- Action collects shared operations: data_forward, mode, pad, make_batch
- Trainer and Tester receives Action as a parameter
- seq_labeling works in such setting
 2018-08-03 21:29:07 +08:00
Coet
8f2d5f7170
Create __init__.py 2018-08-01 20:47:46 +08:00
FengZiYjun
743a6d7547 fix bugs in preprocessor 2018-08-01 10:10:55 +08:00
choosewhatulike
c90e7b0134 Merge branch 'master' of https://github.com/choosewhatulike/fastNLP 2018-08-01 00:32:31 +08:00
choosewhatulike
c42487f987 merge master 2018-08-01 00:24:37 +08:00
choosewhatulike
a077703c99 merge master 2018-08-01 00:20:14 +08:00
choosewhatulike
1548ee7ce2 add recall, f1-score 2018-08-01 00:09:09 +08:00
Coet
2c0e931771
Merge pull request #18 from fastnlp/dev/inference 
A lot of changes.
 2018-07-31 19:54:46 +08:00
FengZiYjun
8e3e6d4579 changes to batch methods 
- [action] add k-means bucketing, partition sequences into buckets of nearly the same length
- [trainer] print train loss every 10 steps
- [loader] cws pku loader split sequence longer than max_seq_len into several shorter sequences
 2018-07-31 10:11:44 +08:00
FengZiYjun
9aad8dff6e update 2018-07-30 20:31:48 +08:00
FengZiYjun
58120998c5 update 2018-07-30 18:08:29 +08:00
FengZiYjun
ef8ec3b9e4 add cws train script and corresponding config file 2018-07-30 09:52:46 +08:00
FengZiYjun
233e8328f7 changes to seq label model, 
- [model] optimize cuda support in seq labeling model
- [test] add test data "pku" for chinese word seg
- test_tester.py and test_cws.py is OK to run!
 2018-07-28 20:18:40 +08:00
FengZiYjun
e531908f40 Merge branch 'dev/inference' of https://github.com/fastnlp/fastNLP 2018-07-28 12:06:12 +08:00
FengZiYjun
242e576a30 changes to trainer, tester, preprocessor, etc. 
- [tester][trainer] add cuda support
- [preprocess] fix label2index for padding label seq
- update README.md
- [test] add test_tester.py
- rename "action" to "core"
 2018-07-28 11:57:25 +08:00
Yige XU
aaac041fc3
Update config_loader.py 2018-07-27 17:06:49 +08:00
choosewhatulike
215a67e7af fix optimizer 2018-07-26 22:45:25 +08:00
choosewhatulike
2e96f7308c fix optimizer 2018-07-26 22:38:54 +08:00
FengZiYjun
c83008add9 fastnlp.py works, see test/test_fastNLP.py for high-level API 2018-07-26 22:19:30 +08:00
choosewhatulike
febe27b5bb Merge remote-tracking branch 'origin/master' 2018-07-25 23:41:45 +08:00
choosewhatulike
e535314753 add recall metrics 2018-07-25 23:37:25 +08:00
FengZiYjun
fe17f611b6 changes to preprocessor, trainer, inference & seq modeling 
- [trainer]rename "batchify" to "make_batch" in trainer
- [trainer]pack (batch_x_pad, seq_len) into batch_x in make_batch for seq labeling, because seq length before pad is needed to make masks
- [trainer]unpack it in data_forward
- [model]shorten model definition
- [inference]build inference class. test_POS_pipeline.py is OK to infer
- [preprocessor]handle pickles in a nicer manner
- [FastNLP] add fastNLP.py as high-level API, not finished yet
 2018-07-25 22:01:57 +08:00
choosewhatulike
9701ab2897 add accuracy_score; fix optimizer 2018-07-25 00:14:16 +08:00
FengZiYjun
22d900b7a3 Merge remote-tracking branch 'origin/master' 
# Conflicts:
#	fastNLP/modules/encoder/embedding.py
#	fastNLP/modules/encoder/lstm.py
 2018-07-24 21:14:23 +08:00
Coet
7c2f260a2f
Merge pull request #16 from fastnlp/feature/preprocess 
add default parameters for modules, in order to decrease required params; update POSPreprocessor to return multi-level lists; add metrics.py
 2018-07-24 21:09:54 +08:00
FengZiYjun
eab9473bd4 Merge branch 'feature/preprocess' of https://github.com/fastnlp/fastNLP 2018-07-24 08:40:27 +08:00