HswOAuth/llm_course
19
16
Fork 0
Code Issues 608 Pull Requests Actions Packages Projects Releases Wiki Activity

像截图中微调大模型任务，微调不到10%损失函数就下降那么低了之后就收敛了，我还有必要继续训练还是直接中断进行部署？ #208

New Issue
Closed
opened 2024-10-17 14:17:00 +08:00 by XPH · 2 comments
XPH commented 2024-10-17 14:17:00 +08:00
Copy Link
No description provided.
[PKPP0@F`X$~N~N0IV47)%B.png
75 KiB
[PKPP0@F`X$~N~N0IV47)%B.png
best_newnew commented 2024-10-17 14:29:35 +08:00
Copy Link

我觉得确实是可以考虑直接中断的，如果训练得过多甚至可能会出现过拟合。
但是要小心喔，提前结束训练需要先确定你的模型会在适当的检查点进行保存，然后再在算法中调用相应的检查点结合原来的预训练模型进行部署。

我觉得确实是可以考虑直接中断的，如果训练得过多甚至可能会出现过拟合。 但是要小心喔，提前结束训练需要先确定你的模型会在适当的检查点进行保存，然后再在算法中调用相应的检查点结合原来的预训练模型进行部署。
XPH commented 2024-10-17 15:40:43 +08:00
Author
Copy Link

我觉得确实是可以考虑直接中断的，如果训练得过多甚至可能会出现过拟合。
但是要小心喔，提前结束训练需要先确定你的模型会在适当的检查点进行保存，然后再在算法中调用相应的检查点结合原来的预训练模型进行部署。

好的感谢您的回复，训练的时候每训练50步进行保存的

> 我觉得确实是可以考虑直接中断的，如果训练得过多甚至可能会出现过拟合。 > 但是要小心喔，提前结束训练需要先确定你的模型会在适当的检查点进行保存，然后再在算法中调用相应的检查点结合原来的预训练模型进行部署。 好的感谢您的回复，训练的时候每训练50步进行保存的
XPH closed this issue 2024-10-19 00:03:09 +08:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels   
bug

Something is not working

  
duplicate

This issue or pull request already exists

  
enhancement

New feature

  
help wanted

Need some help

  
invalid

Something is wrong

  
question

More information is needed

  
wontfix

This won't be fixed

No Label
bug
duplicate
enhancement
help wanted
invalid
question
wontfix
Milestone
Clear milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
No Assignees
2 Participants
Notifications
Due Date
The due date is invalid or out of range. Please use the format 'yyyy-mm-dd'.

No due date set.

Dependencies

No dependencies set.

Reference: HswOAuth/llm_course#208
No description provided.
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?