Claudius Ellsel

I guess you have misunderstood the question. I am trying to get a more verbose output. Currently the training accuracy and test loss is not printed and I'd like to...

Hi @ranahanocka, I wrote > Currently the training **accuracy** and test **loss** is not printed You replied > currently prints the training **loss** and testing **accuracy** Those don't contradict each...

Apparently the values are also not available in the Tensorboard data. Just training accuracy and test loss.

Alright. I will look into the code and see if I can get it to work. Will the ways you proposed also add the results to the Tensorboard data?

Nice, I will look into it. I am running this in the cloud on kaggle.com, so localhost won't work for me. But I will see, whether there is an option...

There has been a Tensorboard integration available on Kaggle, unfortunately that does not seem to work currently. Also I prefer using matplotlib, since I also use it for other stuff,...

Yes, I also found a demo and an announcement, but unfortunately that does currently not work with new notebooks: https://www.kaggle.com/product-feedback/89671#764494. However there seems to be a workaround of forking the...

> Is it a problem that I execute the train.py without changing to its directory before? Apparently it is! So that should be sorted out now.

> However there seems to be a workaround of forking the old example that still works. Just for info: That did not work with MeshCNN, might be caused by outdated...

I made some progress today. I can download the /runs folder from kaggle and process it locally. Since i might want to compare several runs, I also used some script...