test/fastNLP
1
0
Fork 0
mirror of https://gitee.com/fastnlp/fastNLP.git synced 2024-11-29 18:59:01 +08:00
Code Issues Actions Packages Projects Releases Wiki Activity
2,458 Commits 3 Branches 13 Tags 25 MiB
f090e4a99c
Commit Graph

2458 Commits

This Branch
This Branch
All Branches
Author SHA1 Message Date
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
FengZiYjun
647c6c637f edit dataset loaders to follow edited preprocessor; add unit tests for dataset loader 2018-07-24 08:40:14 +08:00
FengZiYjun
1c0eac4c82 optimize model definition of seq modeling; test_POS_pipeline ready to run 2018-07-23 22:26:19 +08:00
Yige XU
a422a4e2d6
Update preprocess.py 2018-07-23 00:14:50 +08:00
Coet
7f4e57a1c9
Update metrics.py 2018-07-22 19:17:57 +08:00
FengZiYjun
92be0722dc add init parameters in modules; optimizer Preprocessor; add metrics.py 2018-07-22 17:20:30 +08:00
choosewhatulike
875fdc46a5 add optimizor construct function 2018-07-22 12:21:05 +08:00
FengZiYjun
7557e2d065 fix bugs: 1. action batchify 2. Trainer optimizer position 3. Trainer.pad 2018-07-21 18:21:19 +08:00
Coet
66d380542a
Merge pull request #14 from keezen/master 
Add classification model.
 2018-07-21 17:15:54 +08:00
Ke Zhen
45690d337c merge 2018-07-21 15:25:43 +08:00
Ke Zhen
4a25bdba9c classification 2018-07-21 14:21:57 +08:00
FengZiYjun
843b7c0e7e To Do (save commit) 2018-07-20 20:21:40 +08:00
FengZiYjun
3d234bf5b2 add model selection (best dev) in Trainer: save the best model during validation; add Inference. 2018-07-17 21:05:53 +08:00
FengZiYjun
174ce35fc8 delete unnecessary files 2018-07-16 21:26:42 +08:00
FengZiYjun
301bbdcd1e add accuracy in POS Tester; 
optimize evaluation output in Trainer;
keep POS pipeline (loader + trainer + tester + saver) OK;
add codes borrowed from FudanParser.
 2018-07-16 21:12:46 +08:00
FengZiYjun
621b79ee19 update configLoader to load hyper-parameters from file 2018-07-16 19:53:36 +08:00
FengZiYjun
08b0edaebb Merge remote-tracking branch 'origin/master' 2018-07-16 18:56:20 +08:00
Coet
b3388684d7
Merge pull request #13 from xuyige/master 
update config loader
 2018-07-16 18:55:51 +08:00
Yige XU
91e02a50b5
update config loader 
add ConfigSection class to save config data
 2018-07-16 18:14:51 +08:00
FengZiYjun
eba1d002ca Merge remote-tracking branch 'origin/master' 2018-07-16 11:58:39 +08:00
Coet
f8b09e80c8
Merge pull request #12 from fastnlp/dev/restructure 
restructure module: 4 classes; add modules; move prototype and rename
 2018-07-16 11:57:30 +08:00
FengZiYjun
2df371a0d0 Merge remote-tracking branch 'origin/master' 2018-07-16 11:40:21 +08:00
FengZiYjun
042f63aa46 codebank 2018-07-15 18:39:06 +08:00
FengZiYjun
ec165ce4ac add model saver and loader 2018-07-15 14:59:41 +08:00
Coet
4a82b5d446
Merge pull request #11 from xuyige/master 
update config loader
 2018-07-14 16:11:02 +08:00
Yige XU
7ccec559f3
Add files via upload 2018-07-14 15:42:44 +08:00
Yige XU
7ff29877cd
Merge pull request #2 from fastnlp/master 
1
 2018-07-14 14:48:56 +08:00
FengZiYjun
eb66cbe6c4 restructure module: 4 classes; add modules; move prototype and rename 2018-07-12 21:53:42 +08:00
FengZiYjun
7514be6f30 - add validation loss into trainer.train 
- restructure: move reproduction outside
- add evaluate in tester
 2018-07-11 21:51:35 +08:00
FengZiYjun
a73087e913 refactor Tester; Tester + Trainer for seq modeling work 2018-07-10 22:00:24 +08:00
FengZiYjun
c98d5924b5 sequence labeling ready to Train! 2018-07-10 20:46:35 +08:00
FengZiYjun
83c032df5d fix bug in CRF comments; optimize PyTorch type conversion. 2018-07-10 18:51:42 +08:00
FengZiYjun
4c9c791304 cancel restriction for base model 2018-07-07 16:59:59 +08:00
FengZiYjun
cca276b8c0 - optimize package calling from test files 
- add people.txt in data_for_tests
- To do: incorrect CRF param in POS_pipeline
 2018-07-07 16:57:57 +08:00
FengZiYjun
d7a8217132 finished POSTrainer 2018-07-07 12:10:34 +08:00
Yige XU
b998593bc5
Merge pull request #1 from fastnlp/master 
1
 2018-07-05 13:57:53 +08:00
FengZiYjun
7ea015c0f9 update trainer: loading data with _pickle; add arguments comments. 2018-07-04 23:28:48 +08:00
FengZiYjun
982503d033 optimize code style 2018-07-04 22:56:24 +08:00
Coet
67557222c0
Merge pull request #9 from xuyige/master 
Add files via upload
 2018-07-03 16:44:59 +08:00
Coet
623f8fa1b2
Update base_preprocess.py 2018-07-03 16:42:14 +08:00
Yige XU
9d92c95063
Add files via upload 2018-07-03 16:36:41 +08:00
FengZiYjun
ceffed6a16 update trainer: add sampling and padding in batchify, add pkl loading in prepare_input, check model loss in get_loss 2018-07-03 09:00:29 +08:00
Coet
a99895223c
Merge pull request #8 from keezen/master 
rename and add kmax pooling module
 2018-07-02 19:27:30 +08:00
Zhen Ke
d6187274be
Rename Conv.py to conv.py 2018-07-02 15:16:17 +08:00
Ke Zhen
7b7826544e add kmax pooling module 2018-07-02 15:12:18 +08:00
Ke Zhen
2569c85c8e modify conv and pool module 2018-07-02 14:02:45 +08:00