Petru-Daniel Tudosiu

As per Tensorflow issue [#24496](https://github.com/tensorflow/tensorflow/issues/24496) it seems to be a tensorflow problem. Could you please try and run [this](https://github.com/tensorflow/docs/blob/master/site/en/tutorials/keras/classification.ipynb) tensorflow example and let us know if the same error appears...

I have never encountered your problem. Also, it seems to be Tensorflow & CUDA related more then NiftyNet related, which is also referenced by the fact that it works on...

A new partial limitation has been created by this extension. If ICNR initialization is used then the upsample factor MUST be an integer. Did the modification on the brach with...

I know that for every new feature for which we need a branch we need to create an issue and name the branch based on the issue number. Did I...

@josemariamoreira Did implement a Y-Net as far as I am aware.

Hi @vfdev-5, The whole decision boils down to a design decision. Do you want the user to take care of this specific checkpointing logic or do you want it to...

Hi @vfdev-5, Thanks for moving so quickly. For my personal use, I already coded the Handler so I am ok ;) I just wanted Ignite to be even more user-friendly

@VainF I have based my implementation of the metrics I outlined in the original post on your library. I am to submit a push request to MONAI after I benchmark...

It is better to try and use a smaller Gaussian kernel for the image. Also, padding is not advised since it results in you calculating the metric in undefined areas....

Dear @mdiazmel, I think you can easily give a Dockerfile that requires different licences from the users as input Arguments as long as you do not give a docker image...