AI Rank
EN
← 开源
atulkum

pointer_summarizer

pytorch implementation of "Get To The Point: Summarization with Pointer-Generator Networks"

Model DevelopmentArchitecturePython
在 GitHub 打开
增长势头
+024 小时新增 Star0.0%
909
Star
234
Fork
+0
本周
1
贡献者
创建于 2018-04-16 · 更新于 2026-08-20 · 今日第 12712 名
主要开发者
atulkum

atulkum/pointer_summarizer 排名第 12,712，有 909 个星标。

同类别比较

atulkum/pointer_summarizer 排名第 12,712，此处有 8 个同类条目；NVIDIA-Merlin/Merlin 排名第 12,710。

同类别比较
名称排名指标
stanfordnlp/pyvene12,706
语言:
Python
星标数:
910
分叉数:
117
简介:
Stanford NLP Python library for understanding and improving PyTorch models via interventions
tencent-ailab/hok_env12,707
语言:
Python
星标数:
910
分叉数:
111
简介:
Honor of Kings AI Open Environment of Tencent
thunlp/WebCPM12,708
语言:
HTML
星标数:
910
分叉数:
73
简介:
Official codes for ACL 2023 paper "WebCPM: Interactive Web Search for Chinese Long-form Question Answering"
NVIDIA-Merlin/Merlin12,710
语言:
Python
星标数:
909
分叉数:
131
简介:
NVIDIA Merlin is an open source library providing end-to-end GPU-accelerated recommender systems, from feature engineering and preprocessing to training deep learning models and running inference in production.
genforce/genforce12,714
语言:
Python
星标数:
909
分叉数:
72
简介:
An efficient PyTorch library for deep generative modeling.
mayuelala/FollowYourClick12,717
语言:
Python
星标数:
909
分叉数:
39
简介:
[AAAI 2025] Follow-Your-Click: This repo is the official implementation of "Follow-Your-Click: Open-domain Regional Image Animation via Short Prompts"
andyzeng/3dmatch-toolbox12,726
语言:
C++
星标数:
907
分叉数:
188
简介:
3DMatch - a 3D ConvNet-based local geometric descriptor for aligning 3D meshes and point clouds.
pathak22/context-encoder12,729
语言:
Lua
星标数:
907
分叉数:
199
简介:
[CVPR 2016] Unsupervised Feature Learning by Image Inpainting using GANs
关键参数与规格
关键参数与规格
参数值
语言Python
类别Model Development
子类别Architecture
星标数909
分叉数234
贡献者1
创建日期2018-04-16
更新日期2026-08-20
1 天星标数0
7 天星标数0
排名12,712
上次排名12,622
排名变化-90
7 天星标排名13,573
上次 7 天星标排名13,585
7 天星标排名变化12
总星标排名11,980
上次总星标排名11,957
总星标排名变化-23
国家USA
README 摘录

pytorch implementation of *Get To The Point: Summarization with Pointer-Generator Networks* 1. Train with pointer generation and coverage loss enabled 2. Training with pointer generation enabled 3. Papers using this code Train with pointer generation and coverage loss enabled After training for 100k iterations with coverage loss enabled (batch size 8) Training with pointer generation enabled After training for 500k iterations (batch size 8) How to run training: 1) Follow data generation instruction from 2) Run start_train. sh

, you might need to change some path and parameters in data_util/config. sh Note: In decode mode beam search batch should have only one example replicated to batch size It is tested on pytorch 0. 7 You need to setup pyrouge to get the rouge score Papers using this code: 1) Automatic Program Synthesis of Long Programs with a Learned Garbage Collector ***NeuroIPS 2018*** 2) Automatic Fact-guided Sentence Modification ***AAAI 2020*** 3) Resurrecting Submodularity in Neural Abstractive Summarization

4) StructSum: Summarization via Structured Representations ***EACL 2021*** 5) Concept Pointer Network for Abstractive Summarization ***EMNLP'2019*** 6) PaddlePaddle version 7) VAE-PGN based Abstractive Model in Multi-stage Architecture for Text Summarization ***INLG2019*** 8) Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning ***EMNLP'2019*** 9) Abstractive Spoken Document Summarization using Hierarchical Model with Multi-stage Attention Diversity Optimization ***INTERSPEECH 2020*** 10) Nutribullets Hybrid: Multi-document Health Summarization ***NAACL 2021*** 11) A Corpus of Very Short Scientific Summaries ***CoNLL

2020*** 12) Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View ***AAAI 2021*** 13) CDEvalSumm: An Empirical Study of Cross-Dataset Evaluation for Neural Summarization Systems ***Findings of EMNLP2020*** 14) A Study on Seq2seq for Sentence Compression in Vietnamese ***PACLIC 2020*** 15) Other Roles Matter! Enhancing Role-Oriented Dialogue Summarization via Role Interactions ***ACL 2022***

常见问题

记录了多少星标和分叉？

atulkum/pointer_summarizer 有 909 个星标和 234 个分叉。

仓库排名是多少？

atulkum/pointer_summarizer 排名第 12,712。

记录的语言和类别是什么？

atulkum/pointer_summarizer 有 909 个星标，语言为 Python，类别为 Model Development。

哪个仓库排名最接近？

atulkum/pointer_summarizer 与 NVIDIA-Merlin/Merlin 比较，后者排名第 12,710，有 909 个星标。

README

pytorch implementation of Get To The Point: Summarization with Pointer-Generator Networks

  1. Train with pointer generation and coverage loss enabled
  2. Training with pointer generation enabled
  3. How to run training
  4. Papers using this code

Train with pointer generation and coverage loss enabled

After training for 100k iterations with coverage loss enabled (batch size 8)

ROUGE-1:
rouge_1_f_score: 0.3907 with confidence interval (0.3885, 0.3928)
rouge_1_recall: 0.4434 with confidence interval (0.4410, 0.4460)
rouge_1_precision: 0.3698 with confidence interval (0.3672, 0.3721)

ROUGE-2:
rouge_2_f_score: 0.1697 with confidence interval (0.1674, 0.1720)
rouge_2_recall: 0.1920 with confidence interval (0.1894, 0.1945)
rouge_2_precision: 0.1614 with confidence interval (0.1590, 0.1636)

ROUGE-l:
rouge_l_f_score: 0.3587 with confidence interval (0.3565, 0.3608)
rouge_l_recall: 0.4067 with confidence interval (0.4042, 0.4092)
rouge_l_precision: 0.3397 with confidence interval (0.3371, 0.3420)

Training with pointer generation enabled

After training for 500k iterations (batch size 8)

ROUGE-1:
rouge_1_f_score: 0.3500 with confidence interval (0.3477, 0.3523)
rouge_1_recall: 0.3718 with confidence interval (0.3693, 0.3745)
rouge_1_precision: 0.3529 with confidence interval (0.3501, 0.3555)

ROUGE-2:
rouge_2_f_score: 0.1486 with confidence interval (0.1465, 0.1508)
rouge_2_recall: 0.1573 with confidence interval (0.1551, 0.1597)
rouge_2_precision: 0.1506 with confidence interval (0.1483, 0.1529)

ROUGE-l:
rouge_l_f_score: 0.3202 with confidence interval (0.3179, 0.3225)
rouge_l_recall: 0.3399 with confidence interval (0.3374, 0.3426)
rouge_l_precision: 0.3231 with confidence interval (0.3205, 0.3256)

How to run training:

  1. Follow data generation instruction from https://github.com/abisee/cnn-dailymail
  2. Run start_train.sh, you might need to change some path and parameters in data_util/config.py
  3. For training run start_train.sh, for decoding run start_decode.sh, and for evaluating run run_eval.sh

Note:

Papers using this code:

  1. Automatic Program Synthesis of Long Programs with a Learned Garbage Collector NeuroIPS 2018 https://github.com/amitz25/PCCoder
  2. Automatic Fact-guided Sentence Modification AAAI 2020 https://github.com/darsh10/split_encoder_pointer_summarizer
  3. Resurrecting Submodularity in Neural Abstractive Summarization
  4. StructSum: Summarization via Structured Representations EACL 2021
  5. Concept Pointer Network for Abstractive Summarization EMNLP'2019 https://github.com/wprojectsn/codes
  6. PaddlePaddle version
  7. VAE-PGN based Abstractive Model in Multi-stage Architecture for Text Summarization INLG2019
  8. Clickbait? Sensational Headline Generation with Auto-tuned Reinforcement Learning EMNLP'2019 https://github.com/HLTCHKUST/sensational_headline
  9. Abstractive Spoken Document Summarization using Hierarchical Model with Multi-stage Attention Diversity Optimization INTERSPEECH 2020
  10. Nutribullets Hybrid: Multi-document Health Summarization NAACL 2021
  11. A Corpus of Very Short Scientific Summaries CoNLL 2020
  12. Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View AAAI 2021
  13. CDEvalSumm: An Empirical Study of Cross-Dataset Evaluation for Neural Summarization Systems Findings of EMNLP2020
  14. A Study on Seq2seq for Sentence Compression in Vietnamese PACLIC 2020
  15. Other Roles Matter! Enhancing Role-Oriented Dialogue Summarization via Role Interactions ACL 2022
排名相邻
同类项目