Closed LukeHe closed 6 years ago
@chovanecm , thank you! I know that, but I want to out put the keras output of model.fit function, how I can do that?
By the way, does sacredboard support the latest tensorboard?
According to the documenation of Keras, model.fit
returns
A History object. Its History.history attribute is a record of training loss values and metrics values at successive epochs, as well as validation loss values and validation metrics values (if applicable).
This sounds like a job for the Sacred Metrics API that Sacredboard can display. However, you have to find by yourself how to iterate over the History
object returned by Keras to add it to the Sacred DB.
By the way, does sacredboard support the latest tensorboard?
No as I haven't found time for Sacredboard since I started my current job :-( But I am already considering steps that would allow me to spend some time on the project again.
Hi again. I have fixed running Tensorboard in the latest dev version. But it always starts it on port 6006, so no multiple tensorboards supported (yet) :(
It is already there when you click on an experiment, a detail view should open. Do you have troubles with it?