Using your trained model

[Amir-Reza-Sadri]

I would like to use your trained model over my dataset. Would it be possible to use your DL model as a pre-trained model and do a transfer learning?

[thomaskuestner]

Which pre-trained model do you want to use?

[Amir-Reza-Sadri]

The one for the whole-body adipose tissue segmentation. If possible the weights and network specifications on both TUEF/DZD and NAKO dataset. May I use them in PyTorch?

[thomaskuestner]

You can find the network specifications in the config files pretrained_model/AT_MRGEN_New and pretrained_model/AT_NAKO.
All the networks were trained on this code using Tensorflow, which can not easily be used in Pytorch. If you want to use it in Pytorch you would first need to check that all functions (and parametrizations) are implemented there exactly the same. Please send me an email if you are still interested in the weights of the network.

[Amir-Reza-Sadri]

Hi Thomas, Regarding our last contact in GitHub, I am writing to request the weights of your DL model for the segmentation of adipose tissue. I would greatly appreciate your consideration in this regard. Kind regards.

…

On Mon, Nov 23, 2020 at 6:17 AM Thomas Kuestner ***@***.***> wrote: You can find the network specifications in the config files pretrained_model/AT_MRGEN_New and pretrained_model/AT_NAKO. All the networks were trained on this code using Tensorflow, which can not easily be used in Pytorch. If you want to use it in Pytorch you would first need to check that all functions (and parametrizations) are implemented there exactly the same. Please send me an email if you are still interested in the weights of the network. — You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub <#12 (comment)>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AMCKGYQQYGAOHMFCGDTSNCDSRJAEVANCNFSM4T2FZ52A> .

-- Amir Reza Sadri Biomedical Research Engineer Case Western Reserve University 2103 Cornell Road, Wolstein 6-133, Cleveland, OH 44106 Tel: (216) 315-1264 ars329@case.edu <jta35@case.edu>

[thomaskuestner]

Solved.