关于程序的resume/pretrained

[wxslby]

如果想从某一个iteration继续训练或者在之前的基础上继续训练，应该怎么修改程序？

[wdayang]

你可以在train 函数几行加上test函数里面的加载预训练模型的模块。 获取Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Wxslby ***@***.***> Sent: Wednesday, November 9, 2022 1:36:00 AM To: wdayang/CTformer ***@***.***> Cc: Subscribed ***@***.***> Subject: [wdayang/CTformer] 关于程序的resume/pretrained (Issue #10) 如果想从某一个iteration继续训练或者在之前的基础上继续训练，应该怎么修改程序？ ― Reply to this email directly, view it on GitHub<#10>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AFQCHCQ642GBLCZ74N74I3DWHNA5BANCNFSM6AAAAAAR3CVRRE>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[wdayang]

Train 函数开始训练之前 获取Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Wxslby ***@***.***> Sent: Wednesday, November 9, 2022 1:36:00 AM To: wdayang/CTformer ***@***.***> Cc: Subscribed ***@***.***> Subject: [wdayang/CTformer] 关于程序的resume/pretrained (Issue #10) 如果想从某一个iteration继续训练或者在之前的基础上继续训练，应该怎么修改程序？ ― Reply to this email directly, view it on GitHub<#10>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AFQCHCQ642GBLCZ74N74I3DWHNA5BANCNFSM6AAAAAAR3CVRRE>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[wxslby]

Train 函数开始训练之前 获取Outlook for Androidhttps://aka.ms/AAb9ysg
…
________________________________ From: Wxslby @.> Sent: Wednesday, November 9, 2022 1:36:00 AM To: wdayang/CTformer @.> Cc: Subscribed @.> Subject: [wdayang/CTformer] 关于程序的resume/pretrained (Issue #10) 如果想从某一个iteration继续训练或者在之前的基础上继续训练，应该怎么修改程序？ ― Reply to this email directly, view it on GitHub<#10>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AFQCHCQ642GBLCZ74N74I3DWHNA5BANCNFSM6AAAAAAR3CVRRE. You are receiving this because you are subscribed to this thread.Message ID: @.>
十分感谢您的回复！想问下如果像您在Masked Autoencoders for Low dose CT denoising提到的使用Ten-fold cross-validation，应该如何去做，以及在Table1中SSIM,RMSE中加减的一个值是什么含义？感谢您对初学者的帮助与解答！

[wdayang]

您好，十折交叉只需要在main函数里面不停地修改的test patient 就行。还有那个'±'是多次实验结果的标准差。 获取Outlook for Android<https://aka.ms/AAb9ysg>

…

________________________________ From: Wxslby ***@***.***> Sent: Monday, November 21, 2022 4:50:05 AM To: wdayang/CTformer ***@***.***> Cc: dayang_wang ***@***.***>; Comment ***@***.***> Subject: Re: [wdayang/CTformer] 关于程序的resume/pretrained (Issue #10) Train 函数开始训练之前 获取Outlook for Androidhttps://aka.ms/AAb9ysg …

________________________________ From: Wxslby @.> Sent: Wednesday, November 9, 2022 1:36:00 AM To: wdayang/CTformer @.> Cc: Subscribed @.> Subject: [wdayang/CTformer] 关于程序的resume/pretrained (Issue #10<#10>) 如果想从某一个iteration继续训练或者在之前的基础上继续训练，应该怎么修改程序？ D Reply to this email directly, view it on GitHub<#10<#10>>, or unsubscribehttps://github.com/notifications/unsubscribe-auth/AFQCHCQ642GBLCZ74N74I3DWHNA5BANCNFSM6AAAAAAR3CVRRE. You are receiving this because you are subscribed to this thread.Message ID: @.> 十分感谢您的回复！想问下如果像您在Masked Autoencoders for Low dose CT denoising提到的使用Ten-fold cross-validation，应该如何去做，以及在Table1中SSIM,RMSE中加减的一个值是什么含义？感谢您对初学者的帮助与解答！ [image]<https://user-images.githubusercontent.com/81366534/203019273-fe63a03a-410c-4cff-a8f2-5d4a4142d265.png> ― Reply to this email directly, view it on GitHub<#10 (comment)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/AFQCHCSPUA6QY6GPIIBIIELWJNAU3ANCNFSM6AAAAAAR3CVRRE>. You are receiving this because you commented.Message ID: ***@***.***>