Bert Large Finetuned Squad2

🚀 bert-large-finetuned-squad2

該模型基於bert-large-uncased微調而來，在SQuAD2.0數據集上表現出色，可用於問答任務，為相關領域提供了強大的支持。

🚀 快速開始

本模型基於 bert-large-uncased 構建，並在 SQuAD2.0 數據集上進行了微調。你可以在 這裡（模型） 和 這裡（數據） 找到對應的論文。

✨ 主要特性

• 語言：英文
• 標籤：pytorch、question-answering
• 數據集：squad2
• 評估指標：exact、f1

💻 使用示例

基礎用法

from transformers.pipelines import pipeline

model_name = "phiyodr/bert-large-finetuned-squad2"
nlp = pipeline('question-answering', model=model_name, tokenizer=model_name)
inputs = {
    'question': 'What discipline did Winkelmann create?',
    'context': 'Johann Joachim Winckelmann was a German art historian and archaeologist. He was a pioneering Hellenist who first articulated the difference between Greek, Greco-Roman and Roman art. "The prophet and founding hero of modern archaeology", Winckelmann was one of the founders of scientific archaeology and first applied the categories of style on a large, systematic basis to the history of art. '
}
nlp(inputs)

🔧 技術細節

訓練流程

{
    "base_model": "bert-large-uncased",
    "do_lower_case": True,
    "learning_rate": 3e-5,
    "num_train_epochs": 4,
    "max_seq_length": 384,
    "doc_stride": 128,
    "max_query_length": 64,
    "batch_size": 96 
}

評估結果

• 數據：dev-v2.0.json
• 腳本：evaluate-v2.0.py （原始腳本來自 這裡）

{
  "exact": 76.22336393497852,
  "f1": 79.72527570261339,
  "total": 11873,
  "HasAns_exact": 76.19770580296895,
  "HasAns_f1": 83.21157193271408,
  "HasAns_total": 5928,
  "NoAns_exact": 76.24894869638352,
  "NoAns_f1": 76.24894869638352,
  "NoAns_total": 5945
}