Closed tuwan1214 closed 1 year ago
I used timm == 0.3.2, torch==1.7.0 to do this work. Higher versions also work for testing, but I am not sure if you can reproduce it in the training stage (ImageNet).
Could you tell me the environments you are using?
My environment is timm==0.3.2 torch==1.9 mmcv-full==1.4.0 mmsegmentation==0.13.0 I don't know if this will have any effect on the use of p2t
------------------ 原始邮件 ------------------ 发件人: "yuhuan-wu/P2T" @.>; 发送时间: 2023年9月11日(星期一) 下午5:24 @.>; @.**@.>; 主题: Re: [yuhuan-wu/P2T] resort (Issue #18)
I used timm == 0.3.2, torch==1.7.0 to do this work. Higher versions also work for testing, but I am not sure if you can reproduce it in the training stage (ImageNet).
Could you tell me the environments you are using?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
It seems the environment is OK. If possible, could you paste the error logs here?
The other versions I tried before had some issues to some extent, but now the problem I am encountering is that the pre training file is not a checkpoint file. I will import the downloaded pre training file path, but I still have this issue.
------------------ 原始邮件 ------------------ 发件人: "yuhuan-wu/P2T" @.>; 发送时间: 2023年9月11日(星期一) 晚上6:28 @.>; @.**@.>; 主题: Re: [yuhuan-wu/P2T] resort (Issue #18)
It seems the environment is OK. If possible, could you paste the error logs here?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
I have encountered this problem many times and I am not sure if it is a bug. TypeError: object of type 'int' has no len(). The specific problem is shown in the figure below. I would like to ask if you have encountered similar problems before.
------------------ 原始邮件 ------------------ 发件人: "yuhuan-wu/P2T" @.>; 发送时间: 2023年9月11日(星期一) 晚上6:28 @.>; @.**@.>; 主题: Re: [yuhuan-wu/P2T] resort (Issue #18)
It seems the environment is OK. If possible, could you paste the error logs here?
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
Please specify where the error is. Besides, I did not see the figure you mentioned. If you do not give enough details, I cannot assist to solve the bug further.
Since we do not know where the issue is without any response, we would like to close this issue.
Thank you very much for your willingness to help me. I guess it may be that the multi-card training strategy is used in the source code, and my server is a single card, so I would like to ask you what changes you need to make for single-card training.
------------------ 原始邮件 ------------------ 发件人: "yuhuan-wu/P2T" @.>; 发送时间: 2023年9月19日(星期二) 上午9:17 @.>; @.**@.>; 主题: Re: [yuhuan-wu/P2T] resort (Issue #18)
Since we do not know where the issue is without any response, we would like to close this issue.
— Reply to this email directly, view it on GitHub, or unsubscribe. You are receiving this because you authored the thread.Message ID: @.***>
If you have troubles on adapting P2T to other repositories. I recommend you to see this project: https://github.com/yuhuan-wu/EDN/tree/master
Hello author, I would like to know what the configuration of the P2P environment you are using is because I may encounter some bugs during my own configuration, which is why I want to know