🚀 arinze/address-match-abp-v2
這是一個句子轉換器模型,它能將句子和段落映射到一個64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
當你安裝了句子轉換器後,使用這個模型就變得很簡單:
pip install -U sentence-transformers
然後你可以像這樣使用該模型:
from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]
model = SentenceTransformer('arinze/address-match-abp-v2')
embeddings = model.encode(sentences)
print(embeddings)
📚 詳細文檔
評估結果
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
訓練情況
該模型使用以下參數進行訓練:
數據加載器
sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader
,長度為3125,參數如下:
{'batch_size': 32}
損失函數
sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss
,參數如下:
{'scale': 20.0, 'similarity_fct': 'cos_sim'}
fit() 方法的參數
{
"epochs": 4,
"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": 157,
"weight_decay": 0.01
}
完整模型架構
SentenceTransformer(
(0): Transformer({'max_seq_length': 64, 'do_lower_case': False}) with Transformer model: BertModel
(1): Pooling({'word_embedding_dimension': 384, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})
(2): Dense({'in_features': 384, 'out_features': 64, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)
引用與作者