Sbert Xtremedistil L6 H256 Uncased Mean Cosine H32

🚀 ABrinkmann/sbert_xtremedistil-l6-h256-uncased-mean-cosine-h32

这是一个 sentence-transformers 模型，它可以将句子和段落映射到一个 32 维的密集向量空间，可用于聚类或语义搜索等任务。

🚀 快速开始

若你已安装 sentence-transformers，使用该模型将十分便捷。

📦 安装指南

pip install -U sentence-transformers

💻 使用示例

基础用法

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

model = SentenceTransformer('ABrinkmann/sbert_xtremedistil-l6-h256-uncased-mean-cosine-h32')
embeddings = model.encode(sentences)
print(embeddings)

📚 详细文档

评估结果

若要对该模型进行自动评估，可参考 Sentence Embeddings Benchmark：https://seb.sbert.net

训练

该模型使用以下参数进行训练：

数据加载器

torch.utils.data.dataloader.DataLoader，长度为 251，参数如下：

{'batch_size': 32, '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：

{
    "epochs": 1,
    "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": 26,
    "weight_decay": 0.01
}

完整模型架构

SentenceTransformer(
  (0): Transformer({'max_seq_length': 16, 'do_lower_case': False}) with Transformer model: BertModel 
  (1): Pooling({'word_embedding_dimension': 256, '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': 256, 'out_features': 32, 'bias': True, 'activation_function': 'torch.nn.modules.activation.Tanh'})
)

引用与作者