Replies: 5 comments 9 replies
-
|
解决了吗？我也是同样的问题！！！ |
Beta Was this translation helpful? Give feedback.
-
|
我也是同样的问题！ |
Beta Was this translation helpful? Give feedback.
-
|
一样 |
Beta Was this translation helpful? Give feedback.
-
|
这个要怎么解决啊，救救孩子 |
Beta Was this translation helpful? Give feedback.
-
|
兄弟们，transformers==4.38.1，peft==0.7.1，改一下这两个库的版本，就没有这个问题了 |
Beta Was this translation helpful? Give feedback.
-
|
Thank you |
Beta Was this translation helpful? Give feedback.
-
|
adapter_config.json 这个配置文件中的参数从哪里加载进去的，参数报一个TypeError: LoraConfig.init() got an unexpected keyword argument |
Beta Was this translation helpful? Give feedback.
-
|
我直接把他们删了，就可以运行了。不知道会不会影响效果 |
Beta Was this translation helpful? Give feedback.
-
|
楼上的兄弟们，你们有修改什么参数的吗？我使用官方的lora微调，才几十条数据，就报CUDA out of memory，我的是2条16G的4080 |
Beta Was this translation helpful? Give feedback.
-
|
微调的config更新了，你调回去就可以貌似 |
Beta Was this translation helpful? Give feedback.
-
|
可以帮忙对比看看是哪一次修改吗？我看了下lora.yaml文件的修改记录，调回去之后还是报oom |
Beta Was this translation helpful? Give feedback.
-
|
|
Beta Was this translation helpful? Give feedback.
-
|
按照这个改回来了还是不行 |
Beta Was this translation helpful? Give feedback.
-
运行代码：python inference_hf.py output\checkpoint-3000/ --prompt "类型#裤风格#英伦风格#简约"
报了如下错误：
RuntimeError: CUDA error: device-side assert triggered
CUDA kernel errors might be asynchronously reported at some other API call, so the stacktrace below might be incorrect.
For debugging consider passing CUDA_LAUNCH_BLOCKING=1.
Compile with
TORCH_USE_CUDA_DSAto enable device-side assertions.Beta Was this translation helpful? Give feedback.
All reactions