🚀 摘要生成模型
这是一个基于GPT2架构的摘要生成模型,可对输入文档进行摘要提取,适用于CNN Daily Mail等数据集。
🚀 快速开始
在右侧面板,你可以试用该模型(不过它仅能处理较短的序列长度)。在右侧面板输入你想要进行摘要提取的文档。
📦 安装指南
该模型(基于GPT2基础架构)可以按以下方式加载:
from transformers import GPT2LMHeadModel, GPT2TokenizerFast
model = GPT2LMHeadModel.from_pretrained("philippelaban/summary_loop46")
tokenizer = GPT2TokenizerFast.from_pretrained("philippelaban/summary_loop46")
💻 使用示例
基础用法
document = "Bouncing Boulders Point to Quakes on Mars. A preponderance of boulder tracks on the red planet may be evidence of recent seismic activity. If a rock falls on Mars, and no one is there to see it, does it leave a trace? Yes, and it's a beautiful herringbone-like pattern, new research reveals. Scientists have now spotted thousands of tracks on the red planet created by tumbling boulders. Delicate chevron-shaped piles of Martian dust and sand frame the tracks, the team showed, and most fade over the course of a few years. Rockfalls have been spotted elsewhere in the solar system, including on the moon and even a comet. But a big open question is the timing of these processes on other worlds — are they ongoing or did they predominantly occur in the past?"
tokenized_document = tokenizer([document], max_length=300, truncation=True, return_tensors="pt")["input_ids"].cuda()
input_shape = tokenized_document.shape
outputs = model.generate(tokenized_document, do_sample=False, max_length=500, num_beams=4, num_return_sequences=4, no_repeat_ngram_size=6, return_dict_in_generate=True, output_scores=True)
candidate_sequences = outputs.sequences[:, input_shape[1]:]
candidate_scores = outputs.sequences_scores.tolist()
for candidate_tokens, score in zip(candidate_sequences, candidate_scores):
summary = tokenizer.decode(candidate_tokens)
print("[Score: %.3f] %s" % (score, summary[:summary.index("END")]))
示例输出
[Score: -0.113] These tracks have been spotted elsewhere in the solar system, including on the red planet, and no one is there to see it, does it leave a trace? Yes, and
[Score: -0.119] Now researchers have spotted thousands of tracks on the red planet created by tumbling boulders in Mars, and no one is there to see it, does it leave a trace?
[Score: -0.214] Here are answers to those questions posed by scientists investigating the tracks discovered by scientists examining the tracks discovered by scientists exploring the tracks discovered by scientists exploring the tracks discovered by scientists exploring the
[Score: -0.388] These are the kinds of questions swirling around whether these tracks exist on Mars, and whether they should be noticed sooner rather than later. Here are some answers: -- The tracks detected
📚 详细文档
你可以在GitHub仓库中获取更多信息、访问评分函数、训练脚本或示例训练日志:https://github.com/CannyLab/summary_loop
📄 许可证
本项目采用Apache 2.0许可证。
属性 |
详情 |
模型类型 |
基于GPT2架构的摘要生成模型 |
训练数据 |
CNN Daily Mail |
许可证 |
Apache 2.0 |