FutureWarning: You are using `torch.load` with `weights_only=False` (the current default value), which uses the default pickle module implicitly. It is possible to construct malicious pickle data which will execute arbitrary code during unpickling (See https://github.com/pytorch/pytorch/blob/main/SECURITY.md#untrusted-models for more details). In a future release, the default value for `weights_only` will be flipped to `True`. This limits the functions that could be executed during unpickling. Arbitrary objects will no longer be allowed to be loaded via this mode unless they are explicitly allowlisted by the user via `torch.serialization.add_safe_globals`. We recommend you start setting `weights_only=True` for any use case where you don't have full control of the loaded file. Please open an issue on GitHub for any issues related to this experimental feature.
Arcmoon-Hu
commented
1 week ago
Reminder
System Info
llamafactory - latest version torch 2.4.0
Reproduction
llamafactory-cli train full/sft
14b模型，8*A800，ds2，neat pack， bs 2，ga 16 初始训练到289step，手动停止，然后从250step续训，训练到252step，oom，稳定复现
Expected behavior
续训应该能正常进行
Others
会和这个告警有关吗