Well, currently, I am not even using trains; I am just trying to figure out what experiment management system my startup should use. If we do use it, I don’t expect we will make heavy contributions to the front-end, but if there was a feature we really wanted, we would certainly consider implementing it and making a PR.

OK, I will look into agents and think about this. One pain we have is that tensorboard logs are stuck on the machine used for training, and we can’t compare models training on two different machines in one tensorboard (unless they mount the same network filesystem). But it is also important to be able to see TB both during training and after it is finished (and even though the log files are large, storage is cheap, so maybe it would be OK to keep them around). I need to think about the bes...

Thanks. That is a good idea to have custom columns, and tags will be very welcome!

We haven’t done anything about it yet. But we are planning to try out a few experiment management systems soon, including trains