如何用使用 p40 训练lora

[hongyix]

Describe the bug

torch._dynamo.exc.BackendCompilerFailed: debug_wrapper raised RuntimeError: Found Tesla P40 which is too old to be supported by the triton GPU compiler, which is used as the backend. Triton only supports devices of CUDA Capability >= 7.0, but your device is of CUDA capability 6.1

To Reproduce

Describe your attempts

• [ ] I checked the documentation and found no answer
• [ ] I checked to make sure that this is not a duplicate issue

按照https://www.reddit.com/r/StableDiffusion/comments/12nh9hn/dreambooth_training_lora_failed_due_to_triton_gpu/ 上说的配置export TORCHDYNAMO_DISABLE=1 重跑后会报

RuntimeError: GET was unable to find an engine to execute this computation

Context

• OS [CentOS Linux]:
• Hardware [Tesla P40]:

[shibing624]

升级cuda

[hongyix]

升级cuda和最新版本的peft就可以了。