Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/xiangking/ark-nlp
A private nlp coding package, which quickly implements the SOTA solutions.
https://github.com/xiangking/ark-nlp
bert nlp transfomer
Last synced: about 2 months ago
JSON representation
A private nlp coding package, which quickly implements the SOTA solutions.
- Host: GitHub
- URL: https://github.com/xiangking/ark-nlp
- Owner: xiangking
- License: apache-2.0
- Created: 2021-02-21T12:55:19.000Z (over 3 years ago)
- Default Branch: main
- Last Pushed: 2023-03-05T08:26:15.000Z (over 1 year ago)
- Last Synced: 2024-07-10T20:27:07.913Z (2 months ago)
- Topics: bert, nlp, transfomer
- Language: Python
- Homepage:
- Size: 876 KB
- Stars: 310
- Watchers: 16
- Forks: 66
- Open Issues: 18
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# ark-nlp
ark-nlp主要是收集和复现学术与工作中常用的NLP模型
## 环境
* python 3
* torch >= 1.0.0, <1.10.0
* tqdm >= 4.56.0
* jieba >= 0.42.1
* transformers >= 3.0.0
* zhon >= 1.1.5
* scipy >= 1.2.0
* scikit-learn >= 0.17.0
## pip安装
```
pip install --upgrade ark-nlp
```
## 项目结构
ark_nlp
开源的自然语言处理库
ark_nlp.dataset
封装数据加载、处理和转化等功能
ark_nlp.nn
封装一些完整的神经网络模型
ark_nlp.processor
封装分词器、词典和构图器等
ark_nlp.factory
封装损失函数、优化器、训练和预测等功能
ark_nlp.model
按实际NLP任务封装常用的模型，方便调用
## 实现的模型
### 预训练模型
| 模型 | 参考文献 |
| :------: | :----------------------------------------------------------: |
| BERT | [BERT:Pre-training of Deep Bidirectional Transformers for Language Understanding](https://arxiv.org/pdf/1810.04805.pdf) |
| ERNIE1.0 | [ERNIE:Enhanced Representation through Knowledge Integration](https://arxiv.org/pdf/1904.09223.pdf) |
| NEZHA | [NEZHA:Neural Contextualized Representation For Chinese Language Understanding](https://arxiv.org/pdf/1909.00204.pdf) |
| Roformer | [Roformer: Enhanced Transformer with Rotary Position Embedding](https://arxiv.org/pdf/2104.09864.pdf) |
| [ERNIE-CTM](https://github.com/xiangking/ark-nlp/blob/main/ark_nlp/nn/layer/ernie_ctm_block.py) | [ERNIE-CTM（ERNIE for Chinese Text Mining）](https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/text_to_knowledge/ernie-ctm) |
### 文本分类 (Text Classification)
| 模型 | 简介 |
| :--------------: | :-----------------------------------------: |
| RNN/CNN/GRU/LSTM | 经典的RNN, CNN, GRU, LSTM等经典文本分类结构 |
| BERT/ERNIE | 常用的预训练模型分类 |
### 文本匹配 (Text Matching)
| 模型 | 简介 |
| :--------: | :------------------------: |
| [BERT/ERNIE](https://github.com/xiangking/ark-nlp/blob/main/test/text_match.ipynb) | 常用的预训练模型匹配分类 |
| [UnsupervisedSimcse](https://github.com/xiangking/ark-nlp/blob/main/test/simcse.ipynb) | 无监督Simcse匹配算法 |
| [CoSENT](https://github.com/xiangking/PyTorch_CoSENT) | [CoSENT：比Sentence-BERT更有效的句向量方案](https://spaces.ac.cn/archives/8847) |
### 命名实体识别 (Named Entity Recognition)
| 模型 | 参考文献 | 论文源码 |
| :--------: | :------------------------: | :-------------------------------------------: |
| [CRF BERT](https://github.com/xiangking/ark-nlp/blob/main/test/crf_bert.ipynb) | | |
| [Biaffine BERT](https://github.com/xiangking/ark-nlp/blob/main/test/biaffine_bert.ipynb) | | |
| [Span BERT](https://github.com/xiangking/ark-nlp/blob/main/test/span_bert.ipynb) | | |
| [Global Pointer BERT](https://github.com/xiangking/ark-nlp/blob/main/test/gobalpoint_bert.ipynb) | [GlobalPointer：用统一的方式处理嵌套和非嵌套NER](https://www.kexue.fm/archives/8373) | |
| [Efficient Global Pointer BERT](https://github.com/xiangking/ark-nlp/blob/main/ark_nlp/model/ner/global_pointer_bert/global_pointer_bert.py) | [Efficient GlobalPointer：少点参数，多点效果](https://spaces.ac.cn/archives/8877) | |
| [W2NER BERT](https://github.com/xiangking/ark-nlp/blob/main/test/w2ner_bert.ipynb) | [Unified Named Entity Recognition as Word-Word Relation Classification](https://arxiv.org/pdf/2112.10070.pdf) | [github](https://github.com/ljynlp/w2ner) |
### 关系抽取 (Relation Extraction)
| 模型 | 参考文献 | 论文源码 |
| :----------------------------------------------------------: | :----------------------------------------------------------: | :-------------------------------------------: |
| [Casrel](https://github.com/xiangking/ark-nlp/blob/main/test/casrel_bert.ipynb) | [A Novel Cascade Binary Tagging Framework for Relational Triple Extraction](https://arxiv.org/pdf/1909.03227.pdf) | [github](https://github.com/weizhepei/CasRel) |
| [PRGC](https://github.com/xiangking/ark-nlp/blob/main/test/prgc_bert.ipynb) | [PRGC: Potential Relation and Global Correspondence Based Joint Relational Triple Extraction](https://arxiv.org/pdf/2106.09895.pdf) | [github](https://github.com/hy-struggle/PRGC) |
### 信息抽取 (Information Extraction)
| 模型 | 参考文献 | 论文源码 |
| :----------------------------------------------------------: | :----------------------------------------------------------: | :-------------------------------------------: |
| [PromptUie](https://github.com/xiangking/ark-nlp/blob/main/example/CMeEE-PromptUIE.ipynb) | [通用信息抽取 UIE(Universal Information Extraction)](https://github.com/PaddlePaddle/PaddleNLP/tree/develop/model_zoo/uie) |[github](https://github.com/PaddlePaddle/PaddleNLP/blob/develop/model_zoo/uie/model.py) |
### 少样本 (Few-Shot Learning)
| 模型 | 参考文献 | 论文源码 |
| :----------------------------------------------------------: | :----------------------------------------------------------: | :-------------------------------------------: |
| [PromptBert](https://github.com/xiangking/ark-nlp/blob/main/test/prompt_bert.ipynb) | [Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing)](https://arxiv.org/abs/2107.13586) ||
## 实际应用
* [CHIP2021-Task3-临床术语标准化任务-第三名](https://github.com/DataArk/CHIP2021-Task3)
* [CHIP2021-Task1-医学对话临床发现阴阳性判别任务-第一名](https://github.com/DataArk/CHIP2021-Task1-Top1)
* [中文医疗信息处理挑战榜CBLUE](https://github.com/DataArk/CBLUE-Baseline)
## 使用例子
完整代码可参考`test`文件夹
* 文本分类
```python
import torch
import pandas as pd
from ark_nlp.model.tc.bert import Bert
from ark_nlp.model.tc.bert import BertConfig
from ark_nlp.model.tc.bert import Dataset
from ark_nlp.model.tc.bert import Task
from ark_nlp.model.tc.bert import get_default_model_optimizer
from ark_nlp.model.tc.bert import Tokenizer
# 加载数据集
# train_data_df的columns必选包含"text"和"label"
# text列为文本，label列为分类标签
tc_train_dataset = Dataset(train_data_df)
tc_dev_dataset = Dataset(dev_data_df)
# 加载分词器
tokenizer = Tokenizer(vocab='nghuyong/ernie-1.0', max_seq_len=30)
# 文本切分、ID化
tc_train_dataset.convert_to_ids(tokenizer)
tc_dev_dataset.convert_to_ids(tokenizer)
# 加载预训练模型
config = BertConfig.from_pretrained('nghuyong/ernie-1.0',
num_labels=len(tc_train_dataset.cat2id))
dl_module = Bert.from_pretrained('nghuyong/ernie-1.0',
config=config)
# 任务构建
num_epoches = 10
batch_size = 32
optimizer = get_default_model_optimizer(dl_module)
model = Task(dl_module, optimizer, 'ce', cuda_device=0)
# 训练
model.fit(tc_train_dataset,
tc_dev_dataset,
lr=2e-5,
epochs=5,
batch_size=batch_size
)
# 推断
from ark_nlp.model.tc.bert import Predictor
tc_predictor_instance = Predictor(model.module, tokenizer, tc_train_dataset.cat2id)
tc_predictor_instance.predict_one_sample(待预测文本)
```
* 文本匹配
```python
import torch
import pandas as pd
from ark_nlp.model.tm.bert import Bert
from ark_nlp.model.tm.bert import BertConfig
from ark_nlp.model.tm.bert import Dataset
from ark_nlp.model.tm.bert import Task
from ark_nlp.model.tm.bert import get_default_model_optimizer
from ark_nlp.model.tm.bert import Tokenizer
# 加载数据集
# train_data_df的columns必选包含"text_a"、"text_b"和"label"
# text_a和text_b列为文本，label列为匹配标签
tm_train_dataset = Dataset(train_data_df)
tm_dev_dataset = Dataset(dev_data_df)
# 加载分词器
tokenizer = Tokenizer(vocab='nghuyong/ernie-1.0', max_seq_len=30)
# 文本切分、ID化
tm_train_dataset.convert_to_ids(tokenizer)
tm_dev_dataset.convert_to_ids(tokenizer)
# 加载预训练模型
config = BertConfig.from_pretrained('nghuyong/ernie-1.0',
num_labels=len(tm_train_dataset.cat2id))
dl_module = Bert.from_pretrained('nghuyong/ernie-1.0',
config=config)
# 任务构建
num_epoches = 10
batch_size = 32
optimizer = get_default_model_optimizer(dl_module)
model = Task(dl_module, optimizer, 'ce', cuda_device=0)
# 训练
model.fit(tm_train_dataset,
tm_dev_dataset,
lr=2e-5,
epochs=5,
batch_size=batch_size
)
# 推断
from ark_nlp.model.tm.bert import Predictor
tm_predictor_instance = Predictor(model.module, tokenizer, tm_train_dataset.cat2id)
tm_predictor_instance.predict_one_sample([待预测文本A, 待预测文本B])
```
* 命名实体
```python
import torch
import pandas as pd
from ark_nlp.model.ner.crf_bert import CRFBert
from ark_nlp.model.ner.crf_bert import CRFBertConfig
from ark_nlp.model.ner.crf_bert import Dataset
from ark_nlp.model.ner.crf_bert import Task
from ark_nlp.model.ner.crf_bert import get_default_model_optimizer
from ark_nlp.model.ner.crf_bert import Tokenizer
# 加载数据集
# train_data_df的columns必选包含"text"和"label"
# text列为文本
# label列为列表形式，列表中每个元素是如下组织的字典
# {'start_idx': 实体首字符在文本的位置, 'end_idx': 实体尾字符在文本的位置, 'type': 实体类型标签, 'entity': 实体}
ner_train_dataset = Dataset(train_data_df)
ner_dev_dataset = Dataset(dev_data_df)
# 加载分词器
tokenizer = Tokenizer(vocab='nghuyong/ernie-1.0', max_seq_len=30)
# 文本切分、ID化
ner_train_dataset.convert_to_ids(tokenizer)
ner_dev_dataset.convert_to_ids(tokenizer)
# 加载预训练模型
config = CRFBertConfig.from_pretrained('nghuyong/ernie-1.0',
num_labels=len(ner_train_dataset.cat2id))
dl_module = CRFBert.from_pretrained('nghuyong/ernie-1.0',
config=config)
# 任务构建
num_epoches = 10
batch_size = 32
optimizer = get_default_model_optimizer(dl_module)
model = Task(dl_module, optimizer, 'ce', cuda_device=0)
# 训练
model.fit(ner_train_dataset,
ner_dev_dataset,
lr=2e-5,
epochs=5,
batch_size=batch_size
)
# 推断
from ark_nlp.model.ner.crf_bert import Predictor
ner_predictor_instance = Predictor(model.module, tokenizer, ner_train_dataset.cat2id)
ner_predictor_instance.predict_one_sample(待抽取文本)
```
* Casrel关系抽取
```python
import torch
import pandas as pd
from ark_nlp.model.re.casrel_bert import CasRelBert
from ark_nlp.model.re.casrel_bert import CasRelBertConfig
from ark_nlp.model.re.casrel_bert import Dataset
from ark_nlp.model.re.casrel_bert import Task
from ark_nlp.model.re.casrel_bert import get_default_model_optimizer
from ark_nlp.model.re.casrel_bert import Tokenizer
from ark_nlp.factory.loss_function import CasrelLoss
# 加载数据集
# train_data_df的columns必选包含"text"和"label"
# text列为文本
# label列为列表形式，列表中每个元素是如下组织的字典
# [头实体, 头实体首字符在文本的位置, 头实体尾字符在文本的位置, 关系类型, 尾实体, 尾实体首字符在文本的位置, 尾实体尾字符在文本的位置]
re_train_dataset = Dataset(train_data_df)
re_dev_dataset = Dataset(dev_data_df,
categories = re_train_dataset.categories,
is_train=False)
# 加载分词器
tokenizer = Tokenizer(vocab='nghuyong/ernie-1.0', max_seq_len=100)
# 文本切分、ID化
# 注意：casrel的代码这部分其实并没有进行切分、ID化，仅是将分词器赋予dataset对象
re_train_dataset.convert_to_ids(tokenizer)
re_dev_dataset.convert_to_ids(tokenizer)
# 加载预训练模型
config = CasRelBertConfig.from_pretrained('nghuyong/ernie-1.0',
num_labels=len(re_train_dataset.cat2id))
dl_module = CasRelBert.from_pretrained('nghuyong/ernie-1.0',
config=config)
# 任务构建
num_epoches = 40
batch_size = 16
optimizer = get_default_model_optimizer(dl_module)
model = Task(dl_module, optimizer, CasrelLoss(), cuda_device=0)
# 训练
model.fit(re_train_dataset,
re_dev_dataset,
lr=2e-5,
epochs=5,
batch_size=batch_size
)
# 推断
from ark_nlp.model.re.casrel_bert import Predictor
casrel_re_predictor_instance = Predictor(model.module, tokenizer, re_train_dataset.cat2id)
casrel_re_predictor_instance.predict_one_sample(待抽取文本)
```
* PRGC关系抽取
```python
import torch
import pandas as pd
from ark_nlp.model.re.prgc_bert import PRGCBert
from ark_nlp.model.re.prgc_bert import PRGCBertConfig
from ark_nlp.model.re.prgc_bert import Dataset
from ark_nlp.model.re.prgc_bert import Task
from ark_nlp.model.re.prgc_bert import get_default_model_optimizer
from ark_nlp.model.re.prgc_bert import Tokenizer
# 加载数据集
# train_data_df的columns必选包含"text"和"label"
# text列为文本
# label列为列表形式，列表中每个元素是如下组织的字典
# [头实体, 头实体首字符在文本的位置, 头实体尾字符在文本的位置, 关系类型, 尾实体, 尾实体首字符在文本的位置, 尾实体尾字符在文本的位置]
re_train_dataset = Dataset(train_df, is_retain_dataset=True)
re_dev_dataset = Dataset(dev_df,
categories = re_train_dataset.categories,
is_train=False)
# 加载分词器
tokenizer = Tokenizer(vocab='nghuyong/ernie-1.0', max_seq_len=100)
# 文本切分、ID化
re_train_dataset.convert_to_ids(tokenizer)
re_dev_dataset.convert_to_ids(tokenizer)
# 加载预训练模型
config = PRGCBertConfig.from_pretrained('nghuyong/ernie-1.0',
num_labels=len(re_train_dataset.cat2id))
dl_module = PRGCBert.from_pretrained('nghuyong/ernie-1.0',
config=config)
# 任务构建
num_epoches = 40
batch_size = 16
optimizer = get_default_model_optimizer(dl_module)
model = Task(dl_module, optimizer, None, cuda_device=0)
# 训练
model.fit(re_train_dataset,
re_dev_dataset,
lr=2e-5,
epochs=5,
batch_size=batch_size
)
# 推断
from ark_nlp.model.re.prgc_bert import Predictor
prgc_re_predictor_instance = Predictor(model.module, tokenizer, re_train_dataset.cat2id)
prgc_re_predictor_instance.predict_one_sample(待抽取文本)
```
## DisscussionGroup
- 公众号：**DataArk**
- wechat ID: **fk95624**
## Main contributors
## Acknowledge
本项目用于收集和复现学术与工作中常用的NLP模型，整合成方便调用的形式，所以参考借鉴了网上很多开源实现，如有不当的地方，还请联系批评指教。
在此，感谢大佬们的开源实现。
