Patentsberta V2

🚀 PatentSBERTa_V2

PatentSBERTa是一個基於深度自然語言處理（NLP）的混合模型，利用增強的SBERT進行專利距離計算和分類。它由奧爾堡大學商學院的AI: Growth - Lab開發。

該模型可將句子和段落映射到768維的密集向量空間，適用於聚類或語義搜索等任務。

• 論文鏈接：https://www.sciencedirect.com/science/article/abs/pii/S0040162524003329
• 代碼倉庫：https://github.com/AI - Growth - Lab/PatentSBERTa

🚀 快速開始

這是一個 sentence - transformers 模型，若要使用該模型，可按照以下步驟操作。

📦 安裝指南

若已安裝 sentence - transformers，使用此模型會很方便，可通過以下命令進行安裝：

pip install -U sentence-transformers

💻 使用示例

基礎用法（Sentence - Transformers）

安裝好 sentence - transformers 後，可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('{MODEL_NAME}')
embeddings = model.encode(sentences)
print(embeddings)

高級用法（HuggingFace Transformers）

若未安裝 sentence - transformers，可按以下方式使用模型：首先將輸入傳遞給Transformer模型，然後對上下文詞嵌入應用正確的池化操作。

from transformers import AutoTokenizer, AutoModel
import torch


#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)


# 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('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')

# 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, mean pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

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

📚 詳細文檔

引用與作者

若引用此模型，請使用以下文獻： Bekamiri, H., Hain, D. S., & Jurowetzki, R. (2024). PatentSBERTa: A deep NLP based hybrid model for patent distance and classification using augmented SBERT. Technological Forecasting and Social Change, 206, 123536.

評估結果

有關此模型的自動評估，請參閱 Sentence Embeddings Benchmark：https://seb.sbert.net

訓練參數

該模型的訓練參數如下：

數據加載器

torch.utils.data.dataloader.DataLoader，長度為1658，參數如下：

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

損失函數

sentence_transformers.losses.CosineSimilarityLoss.CosineSimilarityLoss

fit() 方法的參數：

{
    "epochs": 4,
    "evaluation_steps": 1000,
    "evaluator": "sentence_transformers.evaluation.EmbeddingSimilarityEvaluator.EmbeddingSimilarityEvaluator",
    "max_grad_norm": 1,
    "optimizer_class": "<class 'transformers.optimization.AdamW'>",
    "optimizer_params": {
        "lr": 2e-05
    },
    "scheduler": "WarmupLinear",
    "steps_per_epoch": null,
    "warmup_steps": 664,
    "weight_decay": 0.01
}

完整模型架構

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': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
)

模型標籤與示例

• 模型標籤：
  • sentence - transformers
  • feature - extraction
  • sentence - similarity
  • transformers
• 示例：
  • 源句子："A method for wireless charging using magnetic resonance"
  • 對比句子：
    • "Wireless power transfer through inductive coupling"
    • "A new compound for pharmaceutical use in treating diabetes"
    • "A method for data encryption in wireless communication"
  • 示例標題："Patent Similarity and Dissimilarity"