This great tool is worth paying for!

Tagging my colleague @<1529271085315395584:profile|AmusedCat74> who needs this with me 🙂

Amazing, thanks a lot @<1523701087100473344:profile|SuccessfulKoala55> !

Is the doc on GitHub so we can copy that into a PR?

Yes, we love the HPO app, and are using it :)

Thanks. That would be very helpful. Some of our graphs are logged by optimization steps, whereas some by epochs, so having all called "Iterations" is not ideal.

Logging scalars also leverages ClearML automatic logging. One problem is that this automatic logging seems to keep its own internal "iteration" counter for each scalar, as opposed to keeping track of, say, the optimizer's number of steps.
That can be simply fixed on clearML python lib by allowing to set a per-scalar iteration-multiplier.

OK, so no way to have an automatic dispatch to different, correctly-sized instances, it’s only achievable by submitting to different queues?

(actually, that might even be feasible without touching the UI, depending how the plot is rendered, but I'll check)

(do you welcome PRs?)

Should we have run it from the git repo?

We’re on the PRO 🙂

Do Pipelines work with Hyperparameter search, and with single training jobs?

Thanks @<1523701087100473344:profile|SuccessfulKoala55> ! Any inkling on how soon? Is it days, weeks, or months please? 🙂

Tagging my colleague @<1529271085315395584:profile|AmusedCat74> who made that report.

Thanks @<1523701070390366208:profile|CostlyOstrich36> ! I'll do - and might even peek under the hood see if I can make a PR. What's the best repo for that? Is it that of the ClearML python package?

And yes, I was also referring to tasks ran by the Autoscaler (potentially via the HPO) app, too.

Dang, so unlike screenshots, reports do not survive task deletion :/

Can the “multiple agents on a single queue” scenario, combined with the autoscaler, spawn multiple agents on a single EC2 instance, by chance, please? (thinking e.g. 8 agents on a 8xGPU machine)

@<1523701070390366208:profile|CostlyOstrich36> Any idea please? We could use our 8xA100 as 8 workers, for 8 single-gpu jobs running faster than on a single 1xV100 each.

Screenshot incoming

Tagging @<1529271085315395584:profile|AmusedCat74> my colleague with whom we ran into this issue.

What is the best way to achieve that please?

Happy to jump on a call if easier to make sense of it :)

Thanks for the answers 🙂

Yes

From the doc I seemed to find ways to log 2D scatter plots, but not line plots :/ (found)
It also seems simpler to keep the scalar logging structure, but be able to pass a multiplier (reflecting the eval_n_steps in for example Torch Lightning)

Yes, exactly. Here is the logical sense it makes: I have plots where iterations represent different units: for some these plots iterations (call them A) are optimization steps, while for others (call them B) they are evaluation iterations, occuring every N optimization steps. I would like to either:

• Change the X label so these different plots do not have the same label when they represent different things.
• Or, even better, keep the unique "iterations" label but be able to change how I lo...

It was a debugging session. We haven’t yet tried a “Standard” non-debugging clearml session.

No problem 🙂 Once you’ve merged it, what do we need to do to get the updated version please?