K BERT
1.0.0
رمز SORCE ومجموعات البيانات لـ "K-Bert: تمكين تمثيل اللغة مع Graph Knowledge" ، والذي يتم تنفيذه على أساس إطار UER.
أخبار
برمجة:
Python3
Pytorch >= 1.0
argparse == 1.1
google_model.bin من هنا ، وحفظه إلى models/ الدليل.CnDbpedia.spo من هنا ، وحفظه إلى brain/kgs/ الدليل.datasets/ الدليل.شجرة دليل K-Bert:
K-BERT
├── brain
│ ├── config.py
│ ├── __init__.py
│ ├── kgs
│ │ ├── CnDbpedia.spo
│ │ ├── HowNet.spo
│ │ └── Medical.spo
│ └── knowgraph.py
├── datasets
│ ├── book_review
│ │ ├── dev.tsv
│ │ ├── test.tsv
│ │ └── train.tsv
│ ├── chnsenticorp
│ │ ├── dev.tsv
│ │ ├── test.tsv
│ │ └── train.tsv
│ ...
│
├── models
│ ├── google_config.json
│ ├── google_model.bin
│ └── google_vocab.txt
├── outputs
├── uer
├── README.md
├── requirements.txt
├── run_kbert_cls.py
└── run_kbert_ner.py
قم بتشغيل مثال على مراجعة الكتاب مع CNDBPedia:
CUDA_VISIBLE_DEVICES= ' 0 ' nohup python3 -u run_kbert_cls.py
--pretrained_model_path ./models/google_model.bin
--config_path ./models/google_config.json
--vocab_path ./models/google_vocab.txt
--train_path ./datasets/book_review/train.tsv
--dev_path ./datasets/book_review/dev.tsv
--test_path ./datasets/book_review/test.tsv
--epochs_num 5 --batch_size 32 --kg_name CnDbpedia
--output_model_path ./outputs/kbert_bookreview_CnDbpedia.bin
> ./outputs/kbert_bookreview_CnDbpedia.log &نتائج:
Best accuracy in dev : 88.80%
Best accuracy in test: 87.69%
خيارات run_kbert_cls.py :
useage: [--pretrained_model_path] - Path to the pre-trained model parameters.
[--config_path] - Path to the model configuration file.
[--vocab_path] - Path to the vocabulary file.
--train_path - Path to the training dataset.
--dev_path - Path to the validating dataset.
--test_path - Path to the testing dataset.
[--epochs_num] - The number of training epoches.
[--batch_size] - Batch size of the training process.
[--kg_name] - The name of knowledge graph, "HowNet", "CnDbpedia" or "Medical".
[--output_model_path] - Path to the output model.
دقة (DEV/TEST ٪) على مجموعة بيانات مختلفة:
| مجموعة البيانات | Hownet | cndbpedia |
|---|---|---|
| مراجعة الكتاب | 88.75/87.75 | 88.80/87.69 |
| Chnsenticorp | 95.00/95.50 | 94.42/95.25 |
| التسوق | 97.01/96.92 | 96.94/96.73 |
| ويبو | 98.22/98.33 | 98.29/98.33 |
| LCQMC | 88.97/87.14 | 88.91/87.20 |
| Xnli | 77.11/77.07 | 76.99/77.43 |
قم بتشغيل مثال على مجموعة بيانات MSRA_NER مع CNDBPedia:
CUDA_VISIBLE_DEVICES='0' nohup python3 -u run_kbert_ner.py
--pretrained_model_path ./models/google_model.bin
--config_path ./models/google_config.json
--vocab_path ./models/google_vocab.txt
--train_path ./datasets/msra_ner/train.tsv
--dev_path ./datasets/msra_ner/dev.tsv
--test_path ./datasets/msra_ner/test.tsv
--epochs_num 5 --batch_size 16 --kg_name CnDbpedia
--output_model_path ./outputs/kbert_msraner_CnDbpedia.bin
> ./outputs/kbert_msraner_CnDbpedia.log &
نتائج:
The best in dev : precision=0.957, recall=0.962, f1=0.960
The best in test: precision=0.953, recall=0.959, f1=0.956
خيارات run_kbert_ner.py :
useage: [--pretrained_model_path] - Path to the pre-trained model parameters.
[--config_path] - Path to the model configuration file.
[--vocab_path] - Path to the vocabulary file.
--train_path - Path to the training dataset.
--dev_path - Path to the validating dataset.
--test_path - Path to the testing dataset.
[--epochs_num] - The number of training epoches.
[--batch_size] - Batch size of the training process.
[--kg_name] - The name of knowledge graph.
[--output_model_path] - Path to the output model.
النتائج التجريبية على المهام الخاصة بالمجال (الدقة/الاستدعاء/F1 ٪):
| كجم | finance_qa | Law_qa | Finance_ner | الطب |
|---|---|---|---|---|
| Hownet | 0.805/0.888/0.845 | 0.842/0.903/0.871 | 0.860/0.888/0.874 | 0.935/0.939/0.937 |
| CN-DBPedia | 0.814/0.881/0.846 | 0.814/0.942/0.874 | 0.860/0.887/0.873 | 0.935/0.937/0.936 |
| MedicalKg | - | - | - | 0.944/0.943/0.944 |
هذا العمل هو دراسة مشتركة مع دعم جامعة بكين و Tencent Inc.
إذا كنت تستخدم هذا الرمز ، يرجى الاستشهاد بهذه الورقة:
@inproceedings{weijie2019kbert,
title={{K-BERT}: Enabling Language Representation with Knowledge Graph},
author={Weijie Liu, Peng Zhou, Zhe Zhao, Zhiruo Wang, Qi Ju, Haotang Deng, Ping Wang},
booktitle={Proceedings of AAAI 2020},
year={2020}
}