Patentsberta

🚀 PatentSBERTa

PatentSBERTa 是一个基于深度自然语言处理（NLP）的混合模型，用于专利距离计算和分类。它借助增强的 SBERT 技术，能够将句子和段落映射到 768 维的密集向量空间，可应用于聚类、语义搜索等任务。

🚀 快速开始

本模型可以通过 sentence-transformers 库或 HuggingFace Transformers 库使用，下面分别介绍这两种使用方式。

📦 安装指南

若要使用 sentence-transformers 库调用本模型，需先安装该库：

pip install -U sentence-transformers

💻 使用示例

基础用法（Sentence-Transformers）

安装好 sentence-transformers 后，你可以按如下方式使用模型：

from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('AI-Growth-Lab/PatentSBERTa')
embeddings = model.encode(sentences)
print(embeddings)

高级用法（HuggingFace Transformers）

若未安装 sentence-transformers，可以使用 HuggingFace Transformers 库调用模型。首先将输入数据传入 Transformer 模型，然后对上下文词嵌入应用合适的池化操作：

from transformers import AutoTokenizer, AutoModel
import torch


def cls_pooling(model_output, attention_mask):
    return model_output[0][:,0]


# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('AI-Growth-Lab/PatentSBERTa')
model = AutoModel.from_pretrained('AI-Growth-Lab/PatentSBERTa')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)

# Perform pooling. In this case, cls pooling.
sentence_embeddings = cls_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)

📚 详细文档

评估结果

若要对本模型进行自动化评估，请参考 Sentence Embeddings Benchmark：https://seb.sbert.net

训练

本模型的训练参数如下：

数据加载器

torch.utils.data.dataloader.DataLoader，长度为 5，参数如下：

{'batch_size': 16, 'sampler': 'torch.utils.data.sampler.RandomSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}

损失函数

sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss

fit() 方法的参数如下：

{
    "epochs": 1,
    "evaluation_steps": 0,
    "evaluator": "NoneType",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'transformers.optimization.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 100,
    "weight_decay": 0.01
}

完整模型架构

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: MPNetModel 
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': True, 'pooling_mode_mean_tokens': False, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)

引用与作者

若要引用本模型，请使用以下 BibTeX 格式：

@article{bekamiri2021patentsberta,
  title={PatentSBERTa: A Deep NLP based Hybrid Model for Patent Distance and Classification using Augmented SBERT},
  author={Bekamiri, Hamid and Hain, Daniel S and Jurowetzki, Roman},
  journal={arXiv preprint arXiv:2103.11933},
  year={2021}
}

相关链接

• 论文链接：https://arxiv.org/abs/2103.11933
• 项目 GitHub 链接：https://github.com/AI-Growth-Lab/PatentSBERTa