This is a minimal comet example. I'm afraid I don't know what it does under the hood.. There are no callbacks on the metrics tracked in model.fit and yet if you check out your project in the website, your training and validation losses are tracked automatically, live.

https://www.comet.ml/docs/python-sdk/keras/

Thanks ElegantCoyote26 I'll look into it. Seems like someone liked our automagical approach 🙂

Ah, so you're saying I can write a callback for stuff like train_loss , val_loss , etc.

And then you'll hook it

Absolutely, I could try but I'm not sure what it entails...

FYI 🙂
https://www.tensorflow.org/guide/keras/custom_callback

Thank you!

Is this a possible future feature? I have used cometML before and they have this. I'm not sure how they do it though...

Hey AgitatedDove14 , did you get a chance to look at this?

Hi ElegantCoyote26 , yes I did 🙂
It seems cometml puts their default callback logger for you, that's it.

Thanks!!

Yep

ElegantCoyote26 It means we need to have a keras logger that logs everything to trains, then we need to hook it automatically.
Do you feel like PR-ing the logger (the hooking I can take care of 🙂 )?

ElegantCoyote26 point me to where Keras stores the data 🙂
If in the process of integration you had to add a logger/callback to your Keras code, that is the equivalent of using the TB.

Ok gotchu. I'll do that as soon as I can.

You can check the example here, just make sure you add the callback and you are good to go 🙂
https://github.com/allegroai/trains/blob/master/examples/frameworks/keras/keras_tensorboard.py#L107

ElegantCoyote26 I don't think Keras logs it anywhere unless you have TB, so nowhere to take the data from...
In short, yes you have to have TB :)

Hey AgitatedDove14 , thanks for the answer. What does that mean? In any case I think it would be a nice to have feature.