Giter VIP home page Giter VIP logo

Comments (15)

ganzhiruyi avatar ganzhiruyi commented on June 16, 2024

可以先参考这个看看,yuyuan是wenzhong的医疗版本,https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/wenzhong_qa

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

可以先参考这个看看,yuyuan是wenzhong的医疗版本,https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/wenzhong_qa

请问Wenzhong-GPT2-3.5B模型支持使用Hugging face的trainer接口做继续的预训练吗?因为具体的对话案例是金融领域的,希望提供一些未标注语料进行进一步的预训练。还是可以通过你们框架进行进一步预训练?

from fengshenbang-lm.

ganzhiruyi avatar ganzhiruyi commented on June 16, 2024

可以用huggingface的trainer,但是建议还是参考我们的示例进行进一步预训练,也很简单,避免不必要的踩坑。另外根据这个#111
中的测试,单卡显存至少要达到13G。

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

可以用huggingface的trainer,但是建议还是参考我们的示例进行进一步预训练,也很简单,避免不必要的踩坑。另外根据这个#111 中的测试,单卡显存至少要达到13G。

请问大佬能否提供一下进行进一步预训练的示例代码,谢谢大佬

from fengshenbang-lm.

ganzhiruyi avatar ganzhiruyi commented on June 16, 2024

就是这个,finetune QA对和finetune金融语料是一样的
https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/wenzhong_qa

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

就是这个,finetune QA对和finetune金融语料是一样的 https://github.com/IDEA-CCNL/Fengshenbang-LM/tree/main/fengshen/examples/wenzhong_qa

大佬,我是想找您问一下进行进一步预训练的例子,您给我看的是finetune示例,这两个一样嘛?我的进一步预训练的语料是一些未标注的金融领域文本,打算在这之上,进行qa对的fientune

from fengshenbang-lm.

ganzhiruyi avatar ganzhiruyi commented on June 16, 2024

GPT的pretrain和我们示例的finetune都是一样的,都是LM的过程,你可以先看下代码

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

GPT的pretrain和我们示例的finetune都是一样的,都是LM的过程,你可以先看下代码

好嘞,谢谢大佬,本人属于萌新,是刚刚接触torch,封神框架。谢谢大佬的指教,希望以后能进群,请大佬多带带我。

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

大佬,加了vx号,但是半天没有通过验证,想进群和大家一起交流下,初次用pytorch和封神框架,我感觉有点吃力,代码跑步太起来

from fengshenbang-lm.

koking0 avatar koking0 commented on June 16, 2024

大佬,加了vx号,但是半天没有通过验证,想进群和大家一起交流下,初次用pytorch和封神框架,我感觉有点吃力,代码跑步太起来

我们也在微调这个,可以一起交流一下。

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

们也在微调这个,可以一起交流一下。

可以啊,要不加个vx还是怎么的?

from fengshenbang-lm.

koking0 avatar koking0 commented on June 16, 2024

们也在微调这个,可以一起交流一下。

可以啊,要不加个vx还是怎么的?

Alex_996_007,在封神榜开源讨论群里。

from fengshenbang-lm.

ZeyuTeng96 avatar ZeyuTeng96 commented on June 16, 2024

们也在微调这个,可以一起交流一下。

可以啊,要不加个vx还是怎么的?

Alex_996_007,在封神榜开源讨论群里。

加了,大佬通过下

from fengshenbang-lm.

Tron1994 avatar Tron1994 commented on June 16, 2024

可以用huggingface的trainer,但是建议还是参考我们的示例进行进一步预训练,也很简单,避免不必要的踩坑。另外根据这个#111 中的测试,单卡显存至少要达到13G。

请问finetune目前支持模型并行方式吗

from fengshenbang-lm.

koking0 avatar koking0 commented on June 16, 2024

可以用huggingface的trainer,但是建议还是参考我们的示例进行进一步预训练,也很简单,避免不必要的踩坑。另外根据这个#111 中的测试,单卡显存至少要达到13G。

请问finetune目前支持模型并行方式吗

DeepSpeed开了ZeRO-3的话就是模型并行吧。

from fengshenbang-lm.

Related Issues (20)

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.