🚀 {MODEL_NAME}
這是一個句子轉換器模型:它能將句子和段落映射到一個768維的密集向量空間,可用於聚類或語義搜索等任務。
🚀 快速開始
本模型可通過兩種方式使用,分別是藉助sentence-transformers
庫和直接使用HuggingFace Transformers
庫,下面為你詳細介紹。
📦 安裝指南
若要使用sentence-transformers
庫,你需要先安裝它:
pip install -U sentence-transformers
💻 使用示例
基礎用法(Sentence-Transformers)
當你安裝了sentence-transformers庫後,使用該模型會非常簡單:
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庫,你可以按以下方式使用該模型:首先,將輸入數據傳入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
def cls_pooling(model_output, attention_mask):
return model_output[0][:,0]
sentences = ['This is an example sentence', 'Each sentence is converted']
tokenizer = AutoTokenizer.from_pretrained('{MODEL_NAME}')
model = AutoModel.from_pretrained('{MODEL_NAME}')
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')
with torch.no_grad():
model_output = model(**encoded_input)
sentence_embeddings = cls_pooling(model_output, encoded_input['attention_mask'])
print("Sentence embeddings:")
print(sentence_embeddings)
📚 詳細文檔
評估結果
若要對該模型進行自動評估,請參考 句子嵌入基準測試:https://seb.sbert.net
訓練
該模型使用以下參數進行訓練:
數據加載器:
torch.utils.data.dataloader.DataLoader
,長度為140000,參數如下:
{'batch_size': 32, 'sampler': 'torch.utils.data.sampler.SequentialSampler', 'batch_sampler': 'torch.utils.data.sampler.BatchSampler'}
損失函數:
gpl.toolkit.loss.MarginDistillationLoss
fit()
方法的參數:
{
"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": 140000,
"warmup_steps": 1000,
"weight_decay": 0.01
}
完整模型架構
SentenceTransformer(
(0): Transformer({'max_seq_length': 350, 'do_lower_case': False}) with Transformer model: DistilBertModel
(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})
)
引用與作者