Multi gpus resume error

[minhoooo1]

When I use multi gpus to resume ，the following error reports are generated : Expected all tensors to be on the same device, but found at least two devices, cuda:0 and cuda:1! May be I should change map_location ,but where can i change in pytorch-lighning.

[mBerezovskyy]

@minhoooo1 did you find the solution?

[minhoooo1]

Nope. I haven't paid attention to this issue for a long time

…

------------------ 原始邮件 ------------------ 发件人: ***@***.***>; 发送时间: 2024年2月7日(星期三) 上午7:07 收件人: ***@***.***>; 抄送: ***@***.***>; ***@***.***>; 主题: Re: [Lightning-AI/pytorch-lightning] Multi gpus resume error (Discussion #12458) @minhoooo1 did you find the solution? — Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you were mentioned.Message ID: ***@***.***>