Ecosyste.ms: Issues
An open API service for providing issue and pull request metadata for open source projects.
GitHub / charent/ChatLM-mini-Chinese issues and pull requests
#59 - tokenizer训练OOM 。内存60G
Issue -
State: open - Opened by musexiaoluo 3 months ago
- 1 comment
#59 - tokenizer训练OOM 。内存60G
Issue -
State: open - Opened by musexiaoluo 3 months ago
#58 - 数据清洗代码
Issue -
State: open - Opened by Mrkkew 3 months ago
#58 - 数据清洗代码
Issue -
State: open - Opened by Mrkkew 3 months ago
- 1 comment
#57 - 大佬,能不能分享一下清洗后的数据集呀,loss一直在4.0下不来
Issue -
State: open - Opened by KTVICTORY18 4 months ago
#57 - 大佬,能不能分享一下清洗后的数据集呀,loss一直在4.0下不来
Issue -
State: open - Opened by KTVICTORY18 4 months ago
#56 - 运行3.4python ptr_train.py时报错OSError: Can't load tokenizer for 'D:/pycharmenv/ChatLM-mini-Chinese/model_save/'.
Issue -
State: open - Opened by summerFF 4 months ago
- 1 comment
#56 - 运行3.4python ptr_train.py时报错OSError: Can't load tokenizer for 'D:/pycharmenv/ChatLM-mini-Chinese/model_save/'.
Issue -
State: open - Opened by summerFF 4 months ago
- 2 comments
#55 - 3.4预训练运行出现 unsupported operand type(s) 错误,求帮忙
Issue -
State: open - Opened by summerFF 4 months ago
#55 - 3.4预训练运行出现 unsupported operand type(s) 错误,求帮忙
Issue -
State: open - Opened by summerFF 4 months ago
- 1 comment
#54 - 4080显卡,基本跑不了多少数据,过万条训练数据就报错
Issue -
State: open - Opened by iissy 4 months ago
- 4 comments
#54 - 4080显卡,基本跑不了多少数据,过万条训练数据就报错
Issue -
State: open - Opened by iissy 4 months ago
- 6 comments
#53 - tokenizer的字典中有不少token带有下划线,请问这种是什么意思
Issue -
State: closed - Opened by Mactarvish 5 months ago
- 3 comments
#53 - tokenizer的字典中有不少token带有下划线,请问这种是什么意思
Issue -
State: closed - Opened by Mactarvish 5 months ago
- 2 comments
#52 - 可以用a卡训练吗
Issue -
State: closed - Opened by alexhan1012 6 months ago
- 1 comment
#52 - 可以用a卡训练吗
Issue -
State: closed - Opened by alexhan1012 6 months ago
- 1 comment
#51 - 预训练,用了160万数据,共2G句子对,使用A40的48G显存,无论使用1/2/3/4卡,都会报OOM
Issue -
State: closed - Opened by JaymzWang 6 months ago
- 1 comment
#51 - 预训练,用了160万数据,共2G句子对,使用A40的48G显存,无论使用1/2/3/4卡,都会报OOM
Issue -
State: closed - Opened by JaymzWang 6 months ago
- 1 comment
#50 - 这种只能通过问答对的方式,有没有办法MLM的方式学习知识体系。
Issue -
State: closed - Opened by BShark-YB 7 months ago
- 1 comment
#50 - 这种只能通过问答对的方式,有没有办法MLM的方式学习知识体系。
Issue -
State: closed - Opened by BShark-YB 7 months ago
- 1 comment
#49 - 是否考虑将预训练的模型和仅stf后的模型也上传的平台呢
Issue -
State: closed - Opened by seal-wang 7 months ago
- 1 comment
#49 - 是否考虑将预训练的模型和仅stf后的模型也上传的平台呢
Issue -
State: closed - Opened by seal-wang 7 months ago
- 1 comment
#48 - sft_train
Issue -
State: closed - Opened by dbcSep03 7 months ago
- 1 comment
#48 - sft_train
Issue -
State: closed - Opened by dbcSep03 7 months ago
- 1 comment
#47 - Some NCCL operations have failed or timed out.
Issue -
State: open - Opened by dbcSep03 7 months ago
- 5 comments
#47 - Some NCCL operations have failed or timed out.
Issue -
State: open - Opened by dbcSep03 7 months ago
- 6 comments
#46 - 预训练数据集必须是{“prompt”: "response":}的格式么?
Issue -
State: closed - Opened by dbcSep03 8 months ago
- 2 comments
#46 - 预训练数据集必须是{“prompt”: "response":}的格式么?
Issue -
State: closed - Opened by dbcSep03 8 months ago
- 2 comments
#45 - 非常不错的开源项目
Issue -
State: closed - Opened by DataXujing 8 months ago
- 1 comment
#45 - 非常不错的开源项目
Issue -
State: closed - Opened by DataXujing 8 months ago
- 1 comment
#44 - 请问这些预训练数据加起来有多少token呀
Issue -
State: closed - Opened by StarCycle 8 months ago
- 2 comments
#44 - 请问这些预训练数据加起来有多少token呀
Issue -
State: closed - Opened by StarCycle 8 months ago
- 2 comments
#43 - 这个模型好像没有长文对话的能力,该如何训练它让它有这个能力?
Issue -
State: closed - Opened by Liuxinhao12 8 months ago
- 1 comment
#43 - 这个模型好像没有长文对话的能力,该如何训练它让它有这个能力?
Issue -
State: closed - Opened by Liuxinhao12 8 months ago
- 1 comment
#42 - train_3.5M_CN数据处理问题
Issue -
State: closed - Opened by wflying000 8 months ago
- 1 comment
#42 - train_3.5M_CN数据处理问题
Issue -
State: closed - Opened by wflying000 8 months ago
- 1 comment
#41 - 如何加载sft后的模型?
Issue -
State: closed - Opened by Liuxinhao12 8 months ago
- 1 comment
#41 - 如何加载sft后的模型?
Issue -
State: closed - Opened by Liuxinhao12 8 months ago
- 1 comment
#40 - RuntimeError: No executable batch size found, reached zero
Issue -
State: closed - Opened by suiyueyousan 8 months ago
- 2 comments
#40 - RuntimeError: No executable batch size found, reached zero
Issue -
State: closed - Opened by suiyueyousan 8 months ago
- 2 comments
#39 - 考虑出一个支持llama的版本吗
Issue -
State: closed - Opened by leondada 8 months ago
- 1 comment
#39 - 考虑出一个支持llama的版本吗
Issue -
State: closed - Opened by leondada 8 months ago
- 1 comment
#38 - 如何提取中间层的输出?
Issue -
State: closed - Opened by W-void 8 months ago
- 2 comments
#38 - 如何提取中间层的输出?
Issue -
State: closed - Opened by W-void 8 months ago
- 2 comments
#37 - sft微调时报错
Issue -
State: closed - Opened by ama0zarashi 8 months ago
- 4 comments
#37 - sft微调时报错
Issue -
State: closed - Opened by ama0zarashi 8 months ago
- 4 comments
#36 - 用train.py出现shape的mismatch
Issue -
State: closed - Opened by huluk98 9 months ago
- 10 comments
#36 - 用train.py出现shape的mismatch
Issue -
State: closed - Opened by huluk98 9 months ago
- 10 comments
#35 - 微调后预测三元组不正确原因
Issue -
State: closed - Opened by qiutzh 9 months ago
- 5 comments
#35 - 微调后预测三元组不正确原因
Issue -
State: closed - Opened by qiutzh 9 months ago
- 5 comments
#34 - 预训练数据集
Issue -
State: closed - Opened by rabintang 9 months ago
- 2 comments
#34 - 预训练数据集
Issue -
State: closed - Opened by rabintang 9 months ago
- 2 comments
#33 - 项目怎么使用fastchat 进行调试
Issue -
State: closed - Opened by zhilangtaosha 9 months ago
- 1 comment
#33 - 项目怎么使用fastchat 进行调试
Issue -
State: closed - Opened by zhilangtaosha 9 months ago
- 1 comment
#32 - Great Work! Does it support multimodal ability?
Issue -
State: closed - Opened by LianghuiGuo 9 months ago
- 1 comment
#32 - Great Work! Does it support multimodal ability?
Issue -
State: closed - Opened by LianghuiGuo 9 months ago
- 1 comment
#31 - 运行·pre_train报错,TypeError: Accelerator.__init__() got an unexpected keyword argument 'use_seedable_sampler'
Issue -
State: closed - Opened by JaymzWang 9 months ago
- 1 comment
#31 - 运行·pre_train报错,TypeError: Accelerator.__init__() got an unexpected keyword argument 'use_seedable_sampler'
Issue -
State: closed - Opened by JaymzWang 9 months ago
- 1 comment
#30 - 请问数据预处理里面bell_open_source/train_0.8M_CN.json是在哪里下载的呀
Issue -
State: closed - Opened by PshySimon 10 months ago
- 7 comments
#29 - 请问,如果有新的内容需要添加,是否需要全部重新训练?
Issue -
State: closed - Opened by kideve 10 months ago
- 2 comments
#29 - 请问,如果有新的内容需要添加,是否需要全部重新训练?
Issue -
State: closed - Opened by kideve 10 months ago
- 2 comments
#28 - Bump fastapi from 0.105.0 to 0.109.1
Pull Request -
State: closed - Opened by dependabot[bot] 10 months ago
Labels: dependencies
#27 - 多卡情况下,同一份数据集会加载多次吗
Issue -
State: closed - Opened by shinerdeng 10 months ago
- 6 comments
#26 - 大佬请教一下,只做中文RAG的话,这个跟你另外一个phi,哪个效果比较好?
Issue -
State: closed - Opened by xianzhisheng 10 months ago
- 1 comment
#25 - 请教“3.3 Tokenizer训练”如何运行?
Issue -
State: closed - Opened by ybdesire 10 months ago
- 2 comments
#24 - Why do I get stuck loading the dataset after running it
Issue -
State: closed - Opened by anyiz 10 months ago
- 11 comments
#23 - 在 SFT 微调途中出现报错
Issue -
State: closed - Opened by aoguai 10 months ago
- 11 comments
#22 - 有考虑将模型分发的https://modelscope.cn/么?
Issue -
State: closed - Opened by qmjy 10 months ago
- 2 comments
#21 - 使用Lora 和 sft_train.py 训练效果好像没有,有没有好的方法?
Issue -
State: closed - Opened by yugu91 10 months ago
- 7 comments
#20 - readme可以提供下封装了环境加模型的docker镜像吗?
Issue -
State: closed - Opened by zack-sys 10 months ago
- 1 comment
#19 - 是否有计划针对agent函数调用微调
Issue -
State: closed - Opened by lucasjinreal 10 months ago
- 4 comments
#18 - 如果在更好的设备上训练效果区别大吗
Issue -
State: closed - Opened by aiwillcoming 10 months ago
- 1 comment
#17 - 请教一个问题,生成的回复重复
Issue -
State: closed - Opened by shinerdeng 10 months ago
- 2 comments
#16 - 為甚麼我啟動API會出現這個
Issue -
State: closed - Opened by Adolph3671 10 months ago
- 1 comment
#15 - Hello, 第一次使用,请问运行时出现 unsupported operand type(s) for |: 'types.GenericAlias' and 'type' 是什么问题?
Issue -
State: closed - Opened by yugu91 10 months ago
- 2 comments
#14 - 是否可以在服务器上运行?
Issue -
State: closed - Opened by yanyilin3344 10 months ago
- 5 comments
#13 - 基于提供的模型进行sft报错
Issue -
State: closed - Opened by cq1316 10 months ago
- 13 comments
#12 - 清洗好的数据集会开源吗?
Issue -
State: closed - Opened by echo-valor 11 months ago
- 1 comment
#10 - 如何运行呢?
Issue -
State: closed - Opened by meng25meng 11 months ago
- 17 comments
#9 - 关于小模型ChatLM-mini-Chinese 信息抽取的 sft_train.json文件
Issue -
State: closed - Opened by pengcheng-yan 11 months ago
- 4 comments
#8 - merge from dev to main
Pull Request -
State: closed - Opened by charent 11 months ago
#7 - 可以介绍一下不同的任务训练的配置吗
Issue -
State: closed - Opened by PshySimon 11 months ago
- 4 comments
#6 - merge dev to main
Pull Request -
State: closed - Opened by charent 11 months ago
#5 - merge from dev
Pull Request -
State: closed - Opened by charent 11 months ago
#4 - Bump transformers from 4.35.2 to 4.36.0
Pull Request -
State: closed - Opened by dependabot[bot] 11 months ago
Labels: dependencies
#3 - Bump pyarrow from 13.0.0 to 14.0.1
Pull Request -
State: closed - Opened by dependabot[bot] 12 months ago
Labels: dependencies
#2 - add TODO
Pull Request -
State: closed - Opened by charent about 1 year ago
#1 - Bump pyarrow from 13.0.0 to 14.0.1
Pull Request -
State: closed - Opened by dependabot[bot] about 1 year ago
Labels: dependencies