Mrebel Large 32

🚀 RED^FM：过滤式多语言关系抽取数据集

RED^FM是REBEL的多语言版本。它既可以作为一个独立的多语言关系抽取系统使用，也可以作为预训练系统，在多语言关系抽取数据集上进行微调。

mREBEL在ACL 2023论文RED^{FM}: a Filtered and Multilingual Relation Extraction Dataset中被提出。我们展示了一个新的多语言关系抽取数据集，并训练了一个多语言版本的REBEL，它将关系抽取重新定义为一个序列到序列（seq2seq）任务。论文可在此处找到。如果您使用了代码或模型，请在您的论文中引用这项工作：

@inproceedings{huguet-cabot-et-al-2023-redfm-dataset,
    title = "RED$^{\rm FM}$: a Filtered and Multilingual Relation Extraction Dataset",
    author = "Huguet Cabot, Pere-Llu{\'\i}s  and Tedeschi, Simone and Ngonga Ngomo, Axel-Cyrille and
      Navigli, Roberto",
    booktitle = "Proc. of the 61st Annual Meeting of the Association for Computational Linguistics: ACL 2023",
    month = jul,
    year = "2023",
    address = "Toronto, Canada",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/2306.09802",
}

该论文的原始仓库可在此处找到。

请注意，右侧的推理小部件不会输出特殊标记，而这些标记对于区分主语、宾语和关系类型是必要的。有关mREBEL及其预训练数据集的演示，请查看Spaces演示。

🚀 快速开始

支持语言

• 阿拉伯语（ar）
• 加泰罗尼亚语（ca）
• 德语（de）
• 希腊语（el）
• 英语（en）
• 西班牙语（es）
• 法语（fr）
• 印地语（hi）
• 意大利语（it）
• 日语（ja）
• 韩语（ko）
• 荷兰语（nl）
• 波兰语（pl）
• 葡萄牙语（pt）
• 俄语（ru）
• 瑞典语（sv）
• 越南语（vi）
• 中文（zh）

推理小部件示例

✨ 主要特性

• 多语言支持：支持多种语言的关系抽取任务。
• 可独立使用或微调：既可以作为独立系统，也能在多语言关系抽取数据集上微调。
• 序列到序列任务：将关系抽取重新定义为seq2seq任务。

📦 安装指南

文档未提供具体安装步骤，故跳过此章节。

💻 使用示例

基础用法

from transformers import pipeline

triplet_extractor = pipeline('translation_xx_to_yy', model='Babelscape/mrebel-large-32', tokenizer='Babelscape/mrebel-large-32')
# We need to use the tokenizer manually since we need special tokens.
extracted_text = triplet_extractor.tokenizer.batch_decode([triplet_extractor("The Red Hot Chili Peppers were formed in Los Angeles by Kiedis, Flea, guitarist Hillel Slovak and drummer Jack Irons.", decoder_start_token_id=250058, src_lang="en_XX", tgt_lang="<triplet>", return_tensors=True, return_text=False)[0]["translation_token_ids"]]) # change en_XX for the language of the source.
print(extracted_text[0])
# Function to parse the generated text and extract the triplets
def extract_triplets_typed(text):
    triplets = []
    relation = ''
    text = text.strip()
    current = 'x'
    subject, relation, object_, object_type, subject_type = '','','','',''

    for token in text.replace("<s>", "").replace("<pad>", "").replace("</s>", "").replace("tp_XX", "").replace("__en__", "").split():
        if token == "<triplet>" or token == "<relation>":
            current = 't'
            if relation != '':
                triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
                relation = ''
            subject = ''
        elif token.startswith("<") and token.endswith(">"):
            if current == 't' or current == 'o':
                current = 's'
                if relation != '':
                    triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
                object_ = ''
                subject_type = token[1:-1]
            else:
                current = 'o'
                object_type = token[1:-1]
                relation = ''
        else:
            if current == 't':
                subject += ' ' + token
            elif current == 's':
                object_ += ' ' + token
            elif current == 'o':
                relation += ' ' + token
    if subject != '' and relation != '' and object_ != '' and object_type != '' and subject_type != '':
        triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
    return triplets
extracted_triplets = extract_triplets_typed(extracted_text[0])
print(extracted_triplets)

高级用法

from transformers import AutoModelForSeq2SeqLM, AutoTokenizer

def extract_triplets_typed(text):
    triplets = []
    relation = ''
    text = text.strip()
    current = 'x'
    subject, relation, object_, object_type, subject_type = '','','','',''

    for token in text.replace("<s>", "").replace("<pad>", "").replace("</s>", "").replace("tp_XX", "").replace("__en__", "").split():
        if token == "<triplet>" or token == "<relation>":
            current = 't'
            if relation != '':
                triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
                relation = ''
            subject = ''
        elif token.startswith("<") and token.endswith(">"):
            if current == 't' or current == 'o':
                current = 's'
                if relation != '':
                    triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
                object_ = ''
                subject_type = token[1:-1]
            else:
                current = 'o'
                object_type = token[1:-1]
                relation = ''
        else:
            if current == 't':
                subject += ' ' + token
            elif current == 's':
                object_ += ' ' + token
            elif current == 'o':
                relation += ' ' + token
    if subject != '' and relation != '' and object_ != '' and object_type != '' and subject_type != '':
        triplets.append({'head': subject.strip(), 'head_type': subject_type, 'type': relation.strip(),'tail': object_.strip(), 'tail_type': object_type})
    return triplets

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("Babelscape/mrebel-large-32", src_lang="en_XX", tgt_lang="tp_XX") 
# Here we set English ("en_XX") as source language. To change the source language swap the first token of the input for your desired language or change to supported language. For catalan ("ca_XX") or greek ("el_EL") (not included in mBART pretraining) you need a workaround:
# tokenizer._src_lang = "ca_XX"
# tokenizer.cur_lang_code_id = tokenizer.convert_tokens_to_ids("ca_XX")
# tokenizer.set_src_lang_special_tokens("ca_XX")
model = AutoModelForSeq2SeqLM.from_pretrained("Babelscape/mrebel-large-32")
gen_kwargs = {
    "max_length": 256,
    "length_penalty": 0,
    "num_beams": 3,
    "num_return_sequences": 3,
    "forced_bos_token_id": None,
}

# Text to extract triplets from
text = 'The Red Hot Chili Peppers were formed in Los Angeles by Kiedis, Flea, guitarist Hillel Slovak and drummer Jack Irons.'

# Tokenizer text
model_inputs = tokenizer(text, max_length=256, padding=True, truncation=True, return_tensors = 'pt')

# Generate
generated_tokens = model.generate(
    model_inputs["input_ids"].to(model.device),
    attention_mask=model_inputs["attention_mask"].to(model.device),
    decoder_start_token_id = tokenizer.convert_tokens_to_ids("tp_XX"),
    **gen_kwargs,
)

# Extract text
decoded_preds = tokenizer.batch_decode(generated_tokens, skip_special_tokens=False)

# Extract triplets
for idx, sentence in enumerate(decoded_preds):
    print(f'Prediction triplets sentence {idx}')
    print(extract_triplets_typed(sentence))

📚 详细文档

推理注意事项

右侧的推理小部件不会输出区分主语、宾语和关系类型所需的特殊标记。如需查看mREBEL及其预训练数据集的演示，请访问Spaces演示。

语言设置

在使用模型时，可以设置不同的源语言。对于加泰罗尼亚语（"ca_XX"）或希腊语（"el_EL"）（不包含在mBART预训练中），需要进行一些额外的设置：

tokenizer._src_lang = "ca_XX"
tokenizer.cur_lang_code_id = tokenizer.convert_tokens_to_ids("ca_XX")
tokenizer.set_src_lang_special_tokens("ca_XX")

🔧 技术细节

文档未提供具体技术细节，故跳过此章节。

📄 许可证

该模型遵循CC BY - SA 4.0许可证。许可证文本可在此处找到。