bart large cnn samsum

提供商philschmid
分类summarization
许可证mit
下载量106.2K
星标0

简介

这是一个基于 BART-Large 架构并针对 CNN/DailyMail 和 SAMSum 数据集进行微调的文本摘要模型。它专注于将长文章或对话记录精炼为简洁的摘要，能够较好地处理新闻类报道和日常聊天记录。对于中国开发者而言，该模型适合集成到需要自动化提取要点、生成会议纪要或新闻简报的轻量级应用中。由于是专用微调模型，其在摘要任务上的表现比通用大模型更稳定，且部署成本较低，上手难度极低，直接调用推理接口即可使用。

核心亮点

  • 精通新闻与对话摘要，支持长文本压缩
  • 基于 BART 架构，推理速度快且部署轻量
  • 适合构建自动化纪要或内容精炼工具
  • MIT 协议开源，商业集成灵活且无压力

使用方法

安装依赖
# 安装 Hugging Face transformers
pip install transformers torch
SDK 使用
# 使用 transformers 加载模型
from transformers import AutoModel, AutoTokenizer

model = AutoModel.from_pretrained("philschmid/bart-large-cnn-samsum")
tokenizer = AutoTokenizer.from_pretrained("philschmid/bart-large-cnn-samsum")

Hugging Face 下载

我们推荐使用命令行或者 Hugging Face Hub SDK 来进行模型的下载。

操作指引：在下载前，请先通过如下命令安装 huggingface_hub：

操作指引 
pip install -U huggingface_hub

命令行下载

下载完整模型库

下载完整模型库 
huggingface-cli download philschmid/bart-large-cnn-samsum

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例）

下载单个文件到指定本地文件夹（以下载 config.json 到当前路径下 ./dir 目录为例） 
huggingface-cli download philschmid/bart-large-cnn-samsum config.json --local-dir ./dir

更多命令行下载选项，可参见官方文档

SDK 下载

SDK 下载 
# 模型下载
from huggingface_hub import snapshot_download
model_dir = snapshot_download('philschmid/bart-large-cnn-samsum')

Git 下载

请确保 lfs 已经被正确安装

Git 下载 
git lfs install
git clone https://huggingface.co/philschmid/bart-large-cnn-samsum

如果您希望跳过 lfs 大文件下载，可以使用如下命令

跳过 LFS 
GIT_LFS_SKIP_SMUDGE=1 git clone https://huggingface.co/philschmid/bart-large-cnn-samsum

模型文件托管在 Hugging Face Hub，使用 HF CLI / SDK / Git 直接下载，不经过本站。

PyTorch / Transformers 使用

安装 Transformers

安装 Transformers 
pip install -U transformers torch

模型加载和推理

模型加载和推理 
from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained('philschmid/bart-large-cnn-samsum')
tokenizer = AutoTokenizer.from_pretrained('philschmid/bart-large-cnn-samsum')

完整文档

来源: HuggingFace

---
language: en
license: mit
tags:

  • sagemaker

  • bart

  • summarization

datasets:
  • samsum

widget:
  • text: "Jeff: Can I train a \U0001F917 Transformers model on Amazon SageMaker? \n\

Philipp: Sure you can use the new Hugging Face Deep Learning Container. \nJeff:\
\ ok.\nJeff: and how can I get started? \nJeff: where can I find documentation?\
\ \nPhilipp: ok, ok you can find everything here. https://huggingface.co/blog/the-partnership-amazon-sagemaker-and-hugging-face\n"
model-index:
  • name: bart-large-cnn-samsum

results:
- task:
type: summarization
name: Summarization
dataset:
name: 'SAMSum Corpus: A Human-annotated Dialogue Dataset for Abstractive Summarization'
type: samsum
metrics:
- type: rogue-1
value: 42.621
name: Validation ROGUE-1
- type: rogue-2
value: 21.9825
name: Validation ROGUE-2
- type: rogue-l
value: 33.034
name: Validation ROGUE-L
- type: rogue-1
value: 41.3174
name: Test ROGUE-1
- type: rogue-2
value: 20.8716
name: Test ROGUE-2
- type: rogue-l
value: 32.1337
name: Test ROGUE-L
- task:
type: summarization
name: Summarization
dataset:
name: samsum
type: samsum
config: samsum
split: test
metrics:
- type: rouge
value: 41.3282
name: ROUGE-1
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiZTYzNzZkZDUzOWQzNGYxYTJhNGE4YWYyZjA0NzMyOWUzMDNhMmVhYzY1YTM0ZTJhYjliNGE4MDZhMjhhYjRkYSIsInZlcnNpb24iOjF9.OOM6l3v5rJCndmUIJV-2SDh2NjbPo5IgQOSL-Ju1Gwbi1voL5amsDEDOelaqlUBE3n55KkUsMLZhyn66yWxZBQ
- type: rouge
value: 20.8755
name: ROUGE-2
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMWZiODFiYWQzY2NmOTc5YjA3NTI0YzQ1MzQ0ODk2NjgyMmVlMjA5MjZiNTJkMGRmZGEzN2M3MDNkMjkxMDVhYSIsInZlcnNpb24iOjF9.b8cPk2-IL24La3Vd0hhtii4tRXujh5urAwy6IVeTWHwYfXaURyC2CcQOWtlOx5bdO5KACeaJFrFBCGgjk-VGCQ
- type: rouge
value: 32.1353
name: ROUGE-L
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYWNmYzdiYWQ2ZWRkYzRiMGMxNWUwODgwZTdkY2NjZTc1NWE5NTFiMzU0OTU1N2JjN2ExYWQ2NGZkNjk5OTc4YSIsInZlcnNpb24iOjF9.Fzv4p-TEVicljiCqsBJHK1GsnE_AwGqamVmxTPI0WBNSIhZEhliRGmIL_z1pDq6WOzv3GN2YUGvhowU7GxnyAQ
- type: rouge
value: 38.401
name: ROUGE-LSUM
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiNGI4MWY0NWMxMmQ0ODQ5MDhiNDczMDAzYzJkODBiMzgzYWNkMWM2YTZkZDJmNWJiOGQ3MmNjMGViN2UzYWI2ZSIsInZlcnNpb24iOjF9.7lw3h5k5lJ7tYFLZGUtLyDabFYd00l6ByhmvkW4fykocBy9Blyin4tdw4Xps4DW-pmrdMLgidHxBWz5MrSx1Bw
- type: loss
value: 1.4297215938568115
name: loss
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiMzI0ZWNhNDM5YTViZDMyZGJjMDA1ZWFjYzNhOTdlOTFiNzhhMDBjNmM2MjA3ZmRkZjJjMjEyMGY3MzcwOTI2NyIsInZlcnNpb24iOjF9.oNaZsAtUDqGAqoZWJavlcW7PKx1AWsnkbhaQxadpOKk_u7ywJJabvTtzyx_DwEgZslgDETCf4MM-JKitZKjiDA
- type: gen_len
value: 60.0757
name: gen_len
verified: true
verifyToken: eyJhbGciOiJFZERTQSIsInR5cCI6IkpXVCJ9.eyJoYXNoIjoiYTgwYWYwMDRkNTJkMDM5N2I2MWNmYzQ3OWM1NDJmODUyZGViMGE4ZTdkNmIwYWM2N2VjZDNmN2RiMDE4YTYyYiIsInZlcnNpb24iOjF9.PbXTcNYX_SW-BuRQEcqyc21M7uKrOMbffQSAK6k2GLzTVRrzZxsDC57ktKL68zRY8fSiRGsnknOwv-nAR6YBCQ
---

bart-large-cnn-samsum

> If you want to use the model you should try a newer fine-tuned FLAN-T5 version philschmid/flan-t5-base-samsum out socring the BART version with +6 on ROGUE1 achieving 47.24.

TRY philschmid/flan-t5-base-samsum

This model was trained using Amazon SageMaker and the new Hugging Face Deep Learning container.

For more information look at:





Hyperparameters

json
{
    "dataset_name": "samsum",
    "do_eval": true,
    "do_predict": true,
    "do_train": true,
    "fp16": true,
    "learning_rate": 5e-05,
    "model_name_or_path": "facebook/bart-large-cnn",
    "num_train_epochs": 3,
    "output_dir": "/opt/ml/model",
    "per_device_eval_batch_size": 4,
    "per_device_train_batch_size": 4,
    "predict_with_generate": true,
    "seed": 7
}

Usage

python
from transformers import pipeline
summarizer = pipeline("summarization", model="philschmid/bart-large-cnn-samsum")

conversation = '''Jeff: Can I train a 🤗 Transformers model on Amazon SageMaker? 
Philipp: Sure you can use the new Hugging Face Deep Learning Container. 
Jeff: ok.
Jeff: and how can I get started? 
Jeff: where can I find documentation? 
Philipp: ok, ok you can find everything here. https://huggingface.co/blog/the-partnership-amazon-sagemaker-and-hugging-face                                           
'''
summarizer(conversation)

Results

| key | value |
| --- | ----- |
| eval_rouge1 | 42.621 |
| eval_rouge2 | 21.9825 |
| eval_rougeL | 33.034 |
| eval_rougeLsum | 39.6783 |
| test_rouge1 | 41.3174 |
| test_rouge2 | 20.8716 |
| test_rougeL | 32.1337 |
| test_rougeLsum | 38.4149 |

查看官方来源