Konstantinos Kamnitsas

Very belated reply: DM segments the validation images every `numberOfEpochsBetweenFullInferenceOnValImages` epochs (defined in trainConfig). My guess is that the total number of epochs (seems 100 above) was not exactly divideable...

Thank you for the question and answers. Sorry for the late reply, I am currently writing my PhD thesis and all my time goes there. Indeed, the combined solution of...

Looks good Martin, thanks. Yes, with the other model implementations looking good on the DLTK [paper](https://arxiv.org/abs/1711.06853), users could take for granted that this is polished as well. So lets leave...

Hello, thank you for your interest in our work. The careful preprocessing is necessary for things to work out. What part of the anatomy is imaged? E.g. if it is...

A different set of questions, just to double-check: Have you successfully trained the TinyCNN model? What version of tensorflow are you using? (I am wondering if certain TF versions are...

Hello, thanks for your interest! Yes, sorry, I have not been naming the nodes in the graph :-/ What you need for prediction is indeed this tensor created by the...

Hi, Thank you very much for your interest in our work. I am not 100% sure of exactly what you need. If you mean of saving/loading a whole class instance...

Hello, This is very behind. Because of development of Theano stopped, I prioritized moving to a TensorFlow backend, before extending this project further with extensions such as the adversarials. My...

Thanks a lot! Havent had time to update dependencies and address compatibility for a while, been too busy :-/ Will come back to this soon. Leaving this open until I...

(editted, to be more accurate and complete) pred_seg_u_in_roi = pred_seg_u if roi_mask_u is None else pred_seg_u * roi_mask_u : If roi_mask_u is given, this code essentially "masks" the predictions with...