E5 Small Unsupervised

🚀 E5-small-unsupervised

该模型与 e5-small 类似，但未经过有监督的微调。

Text Embeddings by Weakly-Supervised Contrastive Pre-training Liang Wang, Nan Yang, Xiaolong Huang, Binxing Jiao, Linjun Yang, Daxin Jiang, Rangan Majumder, Furu Wei, arXiv 2022

此模型有 12 层，嵌入维度为 384。

🚀 快速开始

本模型可用于对英文文本进行编码，以实现文本相似度计算、信息检索等任务。

💻 使用示例

基础用法

import torch.nn.functional as F

from torch import Tensor
from transformers import AutoTokenizer, AutoModel


def average_pool(last_hidden_states: Tensor,
                 attention_mask: Tensor) -> Tensor:
    last_hidden = last_hidden_states.masked_fill(~attention_mask[..., None].bool(), 0.0)
    return last_hidden.sum(dim=1) / attention_mask.sum(dim=1)[..., None]


# Each input text should start with "query: " or "passage: ".
# For tasks other than retrieval, you can simply use the "query: " prefix.
input_texts = ['query: how much protein should a female eat',
               'query: summit define',
               "passage: As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",
               "passage: Definition of summit for English Language Learners. : 1  the highest point of a mountain : the top of a mountain. : 2  the highest level. : 3  a meeting or series of meetings between the leaders of two or more governments."]

tokenizer = AutoTokenizer.from_pretrained('intfloat/e5-small-unsupervised')
model = AutoModel.from_pretrained('intfloat/e5-small-unsupervised')

# Tokenize the input texts
batch_dict = tokenizer(input_texts, max_length=512, padding=True, truncation=True, return_tensors='pt')

outputs = model(**batch_dict)
embeddings = average_pool(outputs.last_hidden_state, batch_dict['attention_mask'])

# normalize embeddings
embeddings = F.normalize(embeddings, p=2, dim=1)
scores = (embeddings[:2] @ embeddings[2:].T) * 100
print(scores.tolist())

与 Sentence Transformers 结合使用

from sentence_transformers import SentenceTransformer
model = SentenceTransformer('intfloat/e5-small-unsupervised')
input_texts = [
    'query: how much protein should a female eat',
    'query: summit define',
    "passage: As a general guideline, the CDC's average requirement of protein for women ages 19 to 70 is 46 grams per day. But, as you can see from this chart, you'll need to increase that if you're expecting or training for a marathon. Check out the chart below to see how much protein you should be eating each day.",
    "passage: Definition of summit for English Language Learners. : 1  the highest point of a mountain : the top of a mountain. : 2  the highest level. : 3  a meeting or series of meetings between the leaders of two or more governments."
]
embeddings = model.encode(input_texts, normalize_embeddings=True)

安装依赖

pip install sentence_transformers~=2.2.2

📚 详细文档

训练详情

请参考我们的论文 https://arxiv.org/pdf/2212.03533.pdf。

基准评估

请查看 unilm/e5 以复现该模型在 BEIR 和 MTEB benchmark 上的评估结果。

常见问题解答

1. 是否需要在输入文本前添加 "query: " 和 "passage: " 前缀？

是的，模型是按照这种方式进行训练的，否则会导致性能下降。

以下是一些经验法则：

• 对于非对称任务，如开放问答中的段落检索、临时信息检索，应分别使用 "query: " 和 "passage: "。
• 对于对称任务，如语义相似度、释义检索，使用 "query: " 前缀。
• 如果想将嵌入向量用作特征，如线性探测分类、聚类，使用 "query: " 前缀。

2. 为什么我复现的结果与模型卡片中报告的结果略有不同？

不同版本的 transformers 和 pytorch 可能会导致微小但非零的性能差异。

引用

如果您发现我们的论文或模型有帮助，请考虑按以下方式引用：

@article{wang2022text,
  title={Text Embeddings by Weakly-Supervised Contrastive Pre-training},
  author={Wang, Liang and Yang, Nan and Huang, Xiaolong and Jiao, Binxing and Yang, Linjun and Jiang, Daxin and Majumder, Rangan and Wei, Furu},
  journal={arXiv preprint arXiv:2212.03533},
  year={2022}
}

局限性

该模型仅适用于英文文本，长文本将被截断为最多 512 个词元。

📄 许可证

本项目采用 MIT 许可证。