我在多gpu 运行您的examples/ChatGLM2——transformers.ipynb 的训练脚本，报如下错误： RuntimeError: Expec

你试试在env设置成只有一个cuda可见试试。或者也可以把这两行改成False试试。 <a target="_blank" r

Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0! (when checking argument for argument target in method wrapper_CUDA_nll_loss_forward about torchkeras HOT 1 CLOSED

lyhue1991 commented on August 15, 2024

Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0! (when checking argument for argument target in method wrapper_CUDA_nll_loss_forward

from torchkeras.

Comments (1)

lyhue1991 commented on August 15, 2024

你试试在env设置成只有一个cuda可见试试。

或者也可以把这两行改成False试试。

from torchkeras.

Related Issues (20)

国内镜像源还没有3.9.4版本 HOT 1
怎么保存指定帧的参数？
AttributeError: 'ChatGLM' object has no attribute 'top_p' HOT 4
BaiChuan2_13B微调范例，报错“NotImplementedError: Cannot copy out of meta tensor; no data!” HOT 1
AttributeError: 'Parameter' object has no attribute 'weight' HOT 1
你好，我再vscode中进行使用，好像无法支持vsocde，您有别的方案推荐么，使用wandb，服务器无法连接翻墙
建议增加参数
python3.7版本使用accelerate报错
老铁，能不能把代码的注释加上，代码的参数用途都不晓得，读起来需要根据示例看，太费劲了~~。
FileNotFoundError: [Errno 2] No such file or directory: 'qwen7b_multirounds/adapter_model.bin' 如何解决 HOT 8
大佬，能不能增加一下注释，看着有些混乱
#
能否写一个微调VL-Qwen模型的例子
模型好像更新了，无法调用 HOT 2
ValueError: Can't find 'adapter_config.json'
ValueError: Attempting to unscale FP16 gradients. HOT 2
list index out of range HOT 1
调度器学习率的更新是否该在EpochRunner中进行
请问下 pt 模型文件怎么转化成 transformers 文件结构
关于断点重新训练后先前的记录图像会覆盖掉，不能再次基础上进行记录

Expected all tensors to be on the same device, but found at least two devices, cuda:3 and cuda:0! (when checking argument for argument target in method wrapper_CUDA_nll_loss_forward about torchkeras HOT 1 CLOSED

Comments (1)

Related Issues (20)

Recommend Projects

React

Vue.js

Typescript

TensorFlow

Django

Laravel

D3

Recommend Topics

javascript

web

server

Machine learning

Visualization

Game

Recommend Org

Facebook

Microsoft

Google

Alibaba

D3

Tencent