From fa06c270375c47e135fc5a81afa50c8fb979f61a Mon Sep 17 00:00:00 2001 From: RowitZou <1094074685@qq.com> Date: Wed, 21 Aug 2019 13:15:52 +0800 Subject: [PATCH 1/2] Modify README.md --- README.md | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-) diff --git a/README.md b/README.md index 13fabf5..a71d077 100644 --- a/README.md +++ b/README.md @@ -47,11 +47,13 @@ BMES tag scheme, with each character its label for one line. Sentences are split --saved_model saved_model/model_onto4ner \ --saved_set data/onto4ner.cn/saved.dset -## Data Downloads +## Data The pretrained character and word embeddings can be downloaded from [Lattice LSTM](https://github.com/jiesutd/LatticeLSTM). -Datasets including OntoNotes, MSRA, Weibo and Resume are available at Google Drive or Baidu Pan. +Original datasets can be found at [OntoNotes](https://catalog.ldc.upenn.edu/LDC2011T03), [MSRA](http://sighan.cs.uchicago.edu/bakeoff2006/), +[Weibo](https://github.com/hltcoe/golden-horse) and [Resume](https://github.com/jiesutd/LatticeLSTM/tree/master/ResumeNER). +If you want preprocessed datasets that satisfy the input format of our codes, please contact me. ## Pretrained Model Downloads @@ -59,7 +61,7 @@ We also provide pretrained models on the four datasets, which are the same model If you try to retrain models from scratch under the same hyper-parameter settings, you may obtain a sightly lower or higher F1 score than that reported in the paper (in our experiments we selected the model that performed best). -Pretrained models and related hyper-parameter settings are available at Google Drive or Baidu Pan. +Pretrained models and related hyper-parameter settings are available at Google Drive or Baidu Pan (TBC). When running main.py in test mode for pretrained models, you can get the results as follows: @@ -74,7 +76,7 @@ When running main.py in test mode for pretrained models, you can get the results | Resume dev | 94.27 | 94.59 | 94.43 | | Resume test | 95.28 | 95.46 | 95.37 | -## Cite +## Citation @article{gui2019lexicon, title={A Lexicon-Based Graph Neural Network for Chinese NER}, From fc4cad0e3990938f223d68543769df10279bd750 Mon Sep 17 00:00:00 2001 From: RowitZou <1094074685@qq.com> Date: Fri, 23 Aug 2019 17:44:21 +0800 Subject: [PATCH 2/2] Modify README.md --- README.md | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/README.md b/README.md index a71d077..4f05a98 100644 --- a/README.md +++ b/README.md @@ -61,7 +61,7 @@ We also provide pretrained models on the four datasets, which are the same model If you try to retrain models from scratch under the same hyper-parameter settings, you may obtain a sightly lower or higher F1 score than that reported in the paper (in our experiments we selected the model that performed best). -Pretrained models and related hyper-parameter settings are available at Google Drive or Baidu Pan (TBC). +Pretrained models and related hyper-parameter settings are available at [Google Drive](https://drive.google.com/file/d/1KKkCW8WRhgR2P2UbRpNpKyE_RAv1EREv/view?usp=sharing) or [Baidu Pan](https://pan.baidu.com/s/1U89EwnhPpMa4bNrS--u4EA). When running main.py in test mode for pretrained models, you can get the results as follows: