AI Rank
EN
← 开源
kaiyinzhou

BERT-NER

Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）.

Model DevelopmentDomain-specific trainingPython
在 GitHub 打开
增长势头
+024 小时新增 Star0.0%
1.27k
Star
323
Fork
+0
本周
2
贡献者
创建于 2018-11-07 · 更新于 2026-10-08 · 今日第 10872 名
主要开发者
kaiyinzhoulapolonio

kaiyinzhou/BERT-NER 排名第 10,872，有 1,275 个星标。

同类别比较

kaiyinzhou/BERT-NER 排名第 10,872，此处有 8 个同类条目；MilaNLProc/contextualized-topic-models 排名第 10,874。

同类别比较
名称排名指标
aiming-lab/Agent010,859
语言:
Python
星标数:
1,278
分叉数:
149
简介:
[ICML'26 & COLM'26] Agent0 Series: Self-Evolving Agents from Zero Data
SullyChen/Autopilot-TensorFlow10,863
语言:
Python
星标数:
1,277
分叉数:
432
简介:
A TensorFlow implementation of this Nvidia paper: https://arxiv.org/pdf/1604.07316.pdf with some changes
gfo-project/Gradient-Free-Optimizers10,864
语言:
Python
星标数:
1,277
分叉数:
95
简介:
Lightweight optimization with local, global, population-based and sequential techniques across mixed search spaces
jayleicn/animeGAN10,865
语言:
Jupyter Notebook
星标数:
1,277
分叉数:
193
简介:
A simple PyTorch Implementation of Generative Adversarial Networks, focusing on anime face drawing.
MilaNLProc/contextualized-topic-models10,874
语言:
Python
星标数:
1,274
分叉数:
155
简介:
A python package to run contextualized topic modeling. CTMs combine contextualized embeddings (e.g., BERT) with topic models to get coherent topics. Published at EACL and ACL 2021 (Bianchi et al.).
Linketic/CityGaussian10,876
语言:
Jupyter Notebook
星标数:
1,273
分叉数:
111
简介:
[ECCV`24&ICLR`25] CityGaussian Series for High-quality Large-Scale Scene Reconstruction with Gaussians
google-research/deduplicate-text-datasets10,877
语言:
Rust
星标数:
1,273
分叉数:
130
kakaobrain/nerf-factory10,878
语言:
Python
星标数:
1,272
分叉数:
103
简介:
An awesome PyTorch NeRF library
关键参数与规格
关键参数与规格
参数值
语言Python
类别Model Development
子类别Domain-specific training
星标数1,275
分叉数323
贡献者2
创建日期2018-11-07
更新日期2026-10-08
1 天星标数0
7 天星标数0
排名10,872
上次排名10,842
排名变化-30
7 天星标排名12,802
上次 7 天星标排名12,824
7 天星标排名变化22
总星标排名9,227
上次总星标排名9,217
总星标排名变化-10
国家China
README 摘录

For better performance, you can try NLPGNN, see NLPGNN for more details. BERT-NER Version 2 Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）. The original version （see old_version for more detail） contains some hard codes and lacks corresponding annotations,which is inconvenient to understand. So in this updated version,there are some new ideas and tricks （On data Preprocessing and layer design） that can help you quickly implement the fine-tuning model (you just

need to try to modify crf_layer or softmax_layer). Folder Description: Usage: What's in run_ner. sh: **Notice:** cased model was recommened, according to this paper. CoNLL-2003 dataset and perl Script comes from here RESULTS:(On test set) Parameter setting: do_lower_case=False num_train_epochs=4. 0 crf=False Result description: Here i just use the default paramaters, but as Google's paper says a 0. Maybe some tricks need to be added to the above model

.

常见问题

记录了多少星标和分叉？

kaiyinzhou/BERT-NER 有 1,275 个星标和 323 个分叉。

仓库排名是多少？

kaiyinzhou/BERT-NER 排名第 10,872。

记录的语言和类别是什么？

kaiyinzhou/BERT-NER 有 1,275 个星标，语言为 Python，类别为 Model Development。

哪个仓库排名最接近？

kaiyinzhou/BERT-NER 与 MilaNLProc/contextualized-topic-models 比较，后者排名第 10,874，有 1,274 个星标。

README

For better performance, you can try NLPGNN, see NLPGNN for more details.

BERT-NER Version 2

Use Google's BERT for named entity recognition （CoNLL-2003 as the dataset）.

The original version （see old_version for more detail） contains some hard codes and lacks corresponding annotations,which is inconvenient to understand. So in this updated version,there are some new ideas and tricks （On data Preprocessing and layer design） that can help you quickly implement the fine-tuning model (you just need to try to modify crf_layer or softmax_layer).

Folder Description:

BERT-NER
|____ bert                          # need git from [here](https://github.com/google-research/bert)
|____ cased_L-12_H-768_A-12	    # need download from [here](https://storage.googleapis.com/bert_models/2018_10_18/cased_L-12_H-768_A-12.zip)
|____ data		            # train data
|____ middle_data	            # middle data (label id map)
|____ output			    # output (final model, predict results)
|____ BERT_NER.py		    # mian code
|____ conlleval.pl		    # eval code
|____ run_ner.sh    		    # run model and eval result

Usage:

bash run_ner.sh

What's in run_ner.sh:

python BERT_NER.py\
    --task_name="NER"  \
    --do_lower_case=False \
    --crf=False \
    --do_train=True   \
    --do_eval=True   \
    --do_predict=True \
    --data_dir=data   \
    --vocab_file=cased_L-12_H-768_A-12/vocab.txt  \
    --bert_config_file=cased_L-12_H-768_A-12/bert_config.json \
    --init_checkpoint=cased_L-12_H-768_A-12/bert_model.ckpt   \
    --max_seq_length=128   \
    --train_batch_size=32   \
    --learning_rate=2e-5   \
    --num_train_epochs=3.0   \
    --output_dir=./output/result_dir

perl conlleval.pl -d '\t' < ./output/result_dir/label_test.txt

Notice: cased model was recommened, according to this paper. CoNLL-2003 dataset and perl Script comes from here

RESULTS:(On test set)

Parameter setting:

  • do_lower_case=False
  • num_train_epochs=4.0
  • crf=False
accuracy:  98.15%; precision:  90.61%; recall:  88.85%; FB1:  89.72
              LOC: precision:  91.93%; recall:  91.79%; FB1:  91.86  1387
             MISC: precision:  83.83%; recall:  78.43%; FB1:  81.04  668
              ORG: precision:  87.83%; recall:  85.18%; FB1:  86.48  1191
              PER: precision:  95.19%; recall:  94.83%; FB1:  95.01  1311

Result description:

Here i just use the default paramaters, but as Google's paper says a 0.2% error is reasonable(reported 92.4%). Maybe some tricks need to be added to the above model.

reference:

[1] https://arxiv.org/abs/1810.04805

[2] https://github.com/google-research/bert

排名相邻
同类项目