language: fr

🚨 更新： 此检查点已弃用，请改用 https://huggingface.co/almanach/camembert-base 🚨

CamemBERT：一款美味的法语语言模型

简介

CamemBERT 是基于 RoBERTa 模型的最先进法语语言模型。

该模型现已在 Hugging Face 上提供 6 种不同版本，参数数量、预训练数据量及预训练数据源领域各不相同。

更多信息或请求，请访问 Camembert 官网

预训练模型

如何在 HuggingFace 上使用 CamemBERT

加载 CamemBERT 及其子词分词器：

from transformers import CamembertModel, CamembertTokenizer

# 可将 "camembert-base" 替换为表中任意模型，如 "camembert/camembert-large"。
tokenizer = CamembertTokenizer.from_pretrained("camembert/camembert-base-wikipedia-4gb")
camembert = CamembertModel.from_pretrained("camembert/camembert-base-wikipedia-4gb")

camembert.eval()  # 关闭 dropout（或保持训练模式以进行微调）

使用 pipeline 进行掩码填充

from transformers import pipeline 

camembert_fill_mask  = pipeline("fill-mask", model="camembert/camembert-base-wikipedia-4gb", tokenizer="camembert/camembert-base-wikipedia-4gb")
results = camembert_fill_mask("Le camembert est un fromage de <mask>!")
# 结果
#[{'sequence': '<s> Le camembert est un fromage de chèvre!</s>', 'score': 0.4937814474105835, 'token': 19370}, 
#{'sequence': '<s> Le camembert est un fromage de brebis!</s>', 'score': 0.06255942583084106, 'token': 30616}, 
#{'sequence': '<s> Le camembert est un fromage de montagne!</s>', 'score': 0.04340197145938873, 'token': 2364},
# {'sequence': '<s> Le camembert est un fromage de Noël!</s>', 'score': 0.02823255956172943, 'token': 3236}, 
#{'sequence': '<s> Le camembert est un fromage de vache!</s>', 'score': 0.021357402205467224, 'token': 12329}]

从 Camembert 输出中提取上下文嵌入特征

import torch
# 使用 SentencePiece 进行子词分词
tokenized_sentence = tokenizer.tokenize("J'aime le camembert !")
# ['▁J', "'", 'aime', '▁le', '▁ca', 'member', 't', '▁!'] 

# 进行 one-hot 编码并添加特殊起始和结束标记
encoded_sentence = tokenizer.encode(tokenized_sentence)
# [5, 221, 10, 10600, 14, 8952, 10540, 75, 1114, 6]
# 注：可一步完成：tokenize.encode("J'aime le camembert !")

# 将标记作为 torch 张量输入 Camembert（批次维度为 1）
encoded_sentence = torch.tensor(encoded_sentence).unsqueeze(0)
embeddings, _ = camembert(encoded_sentence)
# embeddings.detach()
# embeddings.size torch.Size([1, 10, 768])
#tensor([[[-0.0928,  0.0506, -0.0094,  ..., -0.2388,  0.1177, -0.1302],
#         [ 0.0662,  0.1030, -0.2355,  ..., -0.4224, -0.0574, -0.2802],
#         [-0.0729,  0.0547,  0.0192,  ..., -0.1743,  0.0998, -0.2677],
#         ...,

从所有 Camembert 层提取上下文嵌入特征

from transformers import CamembertConfig
# （需用新配置重新加载模型）
config = CamembertConfig.from_pretrained("camembert/camembert-base-wikipedia-4gb", output_hidden_states=True)
camembert = CamembertModel.from_pretrained("camembert/camembert-base-wikipedia-4gb", config=config)

embeddings, _, all_layer_embeddings = camembert(encoded_sentence)
#  all_layer_embeddings 列表长度 len(all_layer_embeddings) == 13（输入嵌入层 + 12 个自注意力层）
all_layer_embeddings[5]
# 第 5 层上下文嵌入：尺寸 torch.Size([1, 10, 768])
#tensor([[[-0.0059, -0.0227,  0.0065,  ..., -0.0770,  0.0369,  0.0095],
#         [ 0.2838, -0.1531, -0.3642,  ..., -0.0027, -0.8502, -0.7914],
#         [-0.0073, -0.0338, -0.0011,  ...,  0.0533, -0.0250, -0.0061],
#         ...,

作者

CamemBERT 由 Louis Martin*、Benjamin Muller*、Pedro Javier Ortiz Suárez*、Yoann Dupont、Laurent Romary、Éric Villemonte de la Clergerie、Djamé Seddah 和 Benoît Sagot 训练和评估。

引用

如使用我们的工作，请引用：

@inproceedings{martin2020camembert,
  title={CamemBERT: a Tasty French Language Model},
  author={Martin, Louis and Muller, Benjamin and Su{\'a}rez, Pedro Javier Ortiz and Dupont, Yoann and Romary, Laurent and de la Clergerie, {\'E}ric Villemonte and Seddah, Djam{\'e} and Sagot, Beno{\^\i}t},
  booktitle={Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics},
  year={2020}
}