Skip to content

Issues: PaddlePaddle/PaddleNLP

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

380 Open 3,244 Closed
380 Open 3,244 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

[Question]: 昇腾910b运行llama_npu_sft_N1C8.sh时报错 question Further information is requested
#9469 opened Nov 21, 2024 by feria-tu
@lugimzzz
[Question]: 关于FlashMask的相关问题 question Further information is requested
#9459 opened Nov 19, 2024 by complexfilter
@gongel
[Question]: ddparser无法成功调用 question Further information is requested
#9424 opened Nov 14, 2024 by diwuxiaoxu
@gongel
[Question]: 使用Taskflow进行NER时出错 question Further information is requested
#9413 opened Nov 12, 2024 by tszxyuan123
@wawltor
10
[Question]: 是否支持批量化处理 question Further information is requested
#9406 opened Nov 11, 2024 by lidra
@KB-Ding
[Question]: 请问UIE文档信息抽取标注部分，有现成的案例可以分享么，看标注教程示例看的不是很明白，不知道自己标注出来的数据是否有问题，因为标注了40条数据，效果并没有变好 question Further information is requested
#9405 opened Nov 11, 2024 by lidra
@KB-Ding
1
[Bug]: 安装paddlecustom之后再运行paddlenlp会报错 bug Something isn't working
#9403 opened Nov 11, 2024 by programmer-lxj
1 task done
@DrownFish19
9
[Question]: 文本匹配任务中，交换query和title，会得到完全不同的预测结果 question Further information is requested
#9402 opened Nov 11, 2024 by Yitianw
@KB-Ding
[Question]: paddlenlp调用PaddleCustomDevice custom_cpu报错 question Further information is requested
#9388 opened Nov 7, 2024 by programmer-lxj
@ZHUI
[Bug]: 跑llama3-8b的sft微调时，报错 KeyError: 'eval_accuracy' bug Something isn't working
#9386 opened Nov 7, 2024 by hjx620
1 task done
@ZHUI
1
[Bug]: 跑llama3-8b含tp的prefix-tuning微调时报错ValueError bug Something isn't working
#9384 opened Nov 7, 2024 by hjx620
1 task done
@ZHUI
1
[Bug]: 依存句法分析使用出错，提示LAC依赖，但是LAC已经安装 bug Something isn't working
#9378 opened Nov 6, 2024 by yunc0701
1 task done
@wawltor
5
想请教下：chatglm2-6b用paddlenlp的预训练权重怎么在多卡上推理，另外是否支持在昆仑R200的多卡推理? question Further information is requested
#9371 opened Nov 5, 2024 by xc005
@KB-Ding
在多块昆仑R200的服务器跑代码时，怎么指定使用某块R200? others unknown issue type
#9367 opened Nov 5, 2024 by xc005
@KB-Ding
1
[Question]: 训练ernie文本分类，checkpoint文件夹内没有模型，f1 score一直为0，但loss只有0.02，为什么呢 question Further information is requested
#9351 opened Nov 1, 2024 by yuyu990116
@lugimzzz
2
[Question]: Save 0 examples to ./document/data\train.txt. question Further information is requested
#9317 opened Oct 25, 2024 by wastom
@gongel
4
RISC-V others unknown issue type
#9279 opened Oct 16, 2024 by JocelynPanPan
@KB-Ding
[Question]: UIE可否进行文档级别，在不同文本行之间识别关系呢？ question Further information is requested
#9249 opened Oct 11, 2024 by FlyDut
@wawltor
1
[Question]: 昇腾310P3芯片能否运行paddlenlp？ question Further information is requested
#9238 opened Oct 10, 2024 by ly303550688
@KB-Ding
[Question]: ERNIE模型支持512的token长度，那为何预测时要按照512的字符长度截断呢 question Further information is requested
#9236 opened Oct 9, 2024 by Mengyueke
@DesmonDay
3
[Bug]: 修改精调src_length报错 bug Something isn't working
#9232 opened Oct 9, 2024 by littlesmallrookie
1 task done
@DesmonDay
3
昆仑芯R200跑paddlenlp快速开始失败 question Further information is requested
#9227 opened Oct 8, 2024 by Newcomer-CL
@lugimzzz
1
[Question]: 文本匹配的算法有相关的论文吗 question Further information is requested
#9225 opened Oct 2, 2024 by yangguoer
@lugimzzz
paddlenlp模型部署在移动端 others unknown issue type
#9224 opened Oct 2, 2024 by liuxiaotiao
@lugimzzz
[Question]: 请问是否有多语言版本的UTC模型 question Further information is requested
#9212 opened Sep 27, 2024 by TheHonestBob
@DesmonDay
ProTip! Follow long discussions with comments:>50.