Distiluse-m-v2在stsb_multi_mt上微调的用于西班牙语语义文本相似性的模型

这是一个 sentence-transformers 模型（distiluse-base-multilingual-cased-v2）：它将句子和段落映射到一个768维的稠密向量空间，可用于聚类或语义搜索等任务。

使用方法（Sentence-Transformers）

当您安装了 sentence-transformers 后，使用这个模型变得很容易：

pip install -U sentence-transformers

然后您可以像这样使用该模型：

from sentence_transformers import SentenceTransformer
sentences = ["Nerea va a comprar un cuadro usando bitcoins", "Se puede comprar arte con bitcoins"]

model = SentenceTransformer('mrm8488/distiluse-base-multilingual-cased-v2-finetuned-stsb_multi_mt-es')
embeddings = model.encode(sentences)
print(embeddings)

使用方法（HuggingFace Transformers）

如果没有 sentence-transformers ，您可以像这样使用该模型：首先，将输入通过变换器模型，然后您必须在上下文化词嵌入之上应用正确的汇聚操作。

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 = ["Nerea va a comprar un cuadro usando bitcoins", "Se puede comprar arte con bitcoins"]

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('mrm8488/distiluse-base-multilingual-cased-v2-finetuned-stsb_multi_mt-es')
model = AutoModel.from_pretrained('mrm8488/distiluse-base-multilingual-cased-v2-finetuned-stsb_multi_mt-es')

# 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)

如何评估

from datasets import load_dataset
from sentence_transformers import SentenceTransformer, InputExample
from sentence_transformers.evaluation import EmbeddingSimilarityEvaluator

 
test_data = load_dataset('stsb_multi_mt', 'es', split='test')
test_data = test_data.rename_columns({'similarity_score': 'label'})
test_data = test_data.map(lambda x: {'label': x['label'] / 5.0})

samples = []
for sample in test_data:
    samples.append(InputExample(
        texts=[sample['sentence1'], sample['sentence2']],
        label=sample['label']
    ))
    
evaluator = EmbeddingSimilarityEvaluator.from_input_examples(
    samples, write_csv=False
)

model = SentenceTransformer('mrm8488/distiluse-base-multilingual-cased-v2-finetuned-stsb_multi_mt-es')

evaluator(model)

# It outputs: 0.7604056195656299

评估结果

Spearman等级相关系数：0.7604056195656299

对于对该模型的自动评估，请参见句子嵌入基准： https://seb.sbert.net

训练

模型使用以下参数进行训练：

DataLoader：

sentence_transformers.datasets.NoDuplicatesDataLoader.NoDuplicatesDataLoader，长度为906，参数为：

{'batch_size': 16}

损失函数：

sentence_transformers.losses.MultipleNegativesRankingLoss.MultipleNegativesRankingLoss，参数为：

{'scale': 20.0, 'similarity_fct': 'cos_sim'}

fit()方法的参数：

{
    "epochs": 3,
    "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": 271,
    "weight_decay": 0.01
}

完整模型架构

SentenceTransformer(
  (0): Transformer({'max_seq_length': 512, 'do_lower_case': False}) with Transformer model: DistilBertModel 
  (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})
)

引用和作者