E5-large-unsupervised

这个模型与 e5-large 类似，但没有进行监督的微调。

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

该模型有24层，嵌入尺寸为1024。

使用方法

以下是一个示例，用于对MS-MARCO Passage Ranking数据集中的查询和段落进行编码。

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-large-unsupervised')
model = AutoModel.from_pretrained('intfloat/e5-large-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())

训练细节

请参阅我们在 https://arxiv.org/pdf/2212.03533.pdf 的论文。

基准评估

查看 unilm/e5 以重现对 BEIR 和 MTEB benchmark 的评估结果。

对Sentence Transformers的支持

以下是与sentence_transformers一起使用的示例。

from sentence_transformers import SentenceTransformer
model = SentenceTransformer('intfloat/e5-large-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)

软件包要求

pip install sentence_transformers~=2.2.2

贡献者： michaelfeil

常见问题解答

1. 我需要在输入文本中添加前缀“query:”和“passage:”吗？

是的，在训练模型时，这是必需的，否则性能可能会下降。

以下是一些经验法则：

• 对于不对称任务（如开放式QA中的段落检索，特定信息检索），相应地使用“query:”和“passage:”。

• 对于对称任务（如语义相似性，短语检索），使用“query:”前缀。

• 如果要将嵌入作为特征使用（例如线性探测分类，聚类），请使用“query:”前缀。

2. 为什么我的重复结果与模型卡中报告的结果略有不同？

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

Citation

如果您发现我们的论文或模型有帮助，请考虑按照以下方式引用：

@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}
}

限制

该模型仅适用于英文文本。长文本将被截断为最多512个标记。