Qsingle
Qsingle
> > 是不是在保存参数之后的几百个或者多少个步长之后就会出现这种情况？我在我们的实验上面也遇到了 > > save checkpoint 爆host mem然后ray把进程都杀了吗？我也有这个困扰，有什么办法吗？ 这个目前还没有，等我之后检查一下保存相关的代码看看吧，现在比较赶所以来不及。
> Thanks for the PR! We noticed that hf transformer is doing refactoring for the vlm model interface. We will temporarily pause merging effort to avoid too much adhoc model...
> Thank you for your brilliant work! I’m interested in using Gemma3 as the base to perform RL training and I have tested your PR committed codes on my own...
> Thank you for your suggestion! I have lowered down my transformers to 4.51.3 and observed reasonable rollout output by setting `actor_rollout_ref.rollout.load_format=auto`. > > Also, I noticed your latest commit...
Thank you for your advice, we will fix it in next version.
这个问题是由于标签中的类别ID的最大值大于设定的输出类别值(num_classes)，很可能就是跑的数据集里标签出现了255这种，建议检查一下标签，最后抱歉没有及时回复。
This issue is caused by the label file containing a value greater than the `num_classes` you set. You can check it by printing the maximum value of the label file....
Thanks to @thangngoc89
You can downgrade the transformers to 4.51.0 to solve the problem temporarily. There are some conflicts here.
> 同错误，请问你的vllm是多少，我的vllm是0.8.5.post。安装此版本的transformers会出现“ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts. vllm 0.8.5.post1 requires...