Roberta Base Chinese Extractive Qa

🚀 Chinese RoBERTa-Base Model for QA

This model is designed for extractive question answering. It offers a reliable solution for retrieving answers from given contexts, which is highly useful in various natural language processing scenarios.

🚀 Quick Start

The model can be directly used with a pipeline for extractive question answering. Here is an example of how to use it:

>>> from transformers import AutoModelForQuestionAnswering,AutoTokenizer,pipeline
>>> model = AutoModelForQuestionAnswering.from_pretrained('uer/roberta-base-chinese-extractive-qa')
>>> tokenizer = AutoTokenizer.from_pretrained('uer/roberta-base-chinese-extractive-qa')
>>> QA = pipeline('question-answering', model=model, tokenizer=tokenizer)
>>> QA_input = {'question': "著名诗歌《假如生活欺骗了你》的作者是",'context': "普希金从那里学习人民的语言，吸取了许多有益的养料，这一切对普希金后来的创作产生了很大的影响。这两年里，普希金创作了不少优秀的作品，如《囚徒》、《致大海》、《致凯恩》和《假如生活欺骗了你》等几十首抒情诗，叙事诗《努林伯爵》，历史剧《鲍里斯·戈都诺夫》，以及《叶甫盖尼·奥涅金》前六章。"}
>>> QA(QA_input)
    {'score': 0.9766426682472229, 'start': 0, 'end': 3, 'answer': '普希金'}

✨ Features

• Fine-tuning Flexibility: The model can be fine-tuned by UER-py introduced in this paper, and also by TencentPretrain introduced in this paper. TencentPretrain inherits UER-py to support models with parameters above one billion and extends it to a multimodal pre-training framework.
• Multiple Download Sources: You can download the model either from the UER-py Modelzoo page, or via HuggingFace from the link roberta-base-chinese-extractive-qa.

📦 Installation

There is no specific installation steps provided in the original README. If you want to use the model, you can follow the usage example above to load the pre - trained model.

📚 Documentation

Model description

The model is used for extractive question answering. It is fine - tuned by UER-py and TencentPretrain as mentioned above.

Training data

Training data comes from three sources: cmrc2018, webqa, and laisi. Only the train set of these three datasets is used.

Training procedure

The model is fine - tuned by UER-py on Tencent Cloud. It is fine - tuned for three epochs with a sequence length of 512 on the basis of the pre - trained model chinese_roberta_L-12_H-768. At the end of each epoch, the model is saved when the best performance on the development set is achieved.

python3 finetune/run_cmrc.py --pretrained_model_path models/cluecorpussmall_roberta_base_seq512_model.bin-250000 \
                             --vocab_path models/google_zh_vocab.txt \
                             --train_path datasets/extractive_qa.json \
                             --dev_path datasets/cmrc2018/dev.json \
                             --output_model_path models/extractive_qa_model.bin \
                             --learning_rate 3e-5 --epochs_num 3 --batch_size 32 --seq_length 512

Finally, convert the fine - tuned model into Huggingface's format:

python3 scripts/convert_bert_extractive_qa_from_uer_to_huggingface.py --input_model_path models/extractive_qa_model.bin \
                                                                      --output_model_path pytorch_model.bin \
                                                                      --layers_num 12

BibTeX entry and citation info

@article{liu2019roberta,
  title={Roberta: A robustly optimized bert pretraining approach},
  author={Liu, Yinhan and Ott, Myle and Goyal, Naman and Du, Jingfei and Joshi, Mandar and Chen, Danqi and Levy, Omer and Lewis, Mike and Zettlemoyer, Luke and Stoyanov, Veselin},
  journal={arXiv preprint arXiv:1907.11692},
  year={2019}
}

@article{zhao2019uer,
  title={UER: An Open-Source Toolkit for Pre-training Models},
  author={Zhao, Zhe and Chen, Hui and Zhang, Jinbin and Zhao, Xin and Liu, Tao and Lu, Wei and Chen, Xi and Deng, Haotang and Ju, Qi and Du, Xiaoyong},
  journal={EMNLP-IJCNLP 2019},
  pages={241},
  year={2019}
}

@article{zhao2023tencentpretrain,
  title={TencentPretrain: A Scalable and Flexible Toolkit for Pre-training Models of Different Modalities},
  author={Zhao, Zhe and Li, Yudong and Hou, Cheng and Zhao, Jing and others},
  journal={ACL 2023},
  pages={217},
  year={2023}

📄 License

There is no license information provided in the original README, so this section is skipped.

🔧 Technical Details

There is no specific technical details provided in the original README that meet the requirement of more than 50 - word specific technical description, so this section is skipped.