In that case you have the "packages" parameter for both the controller and the steps

Logs shows me that key is mounted to the docker container

How are you mounting the credentials?
What version of ClearML-Agent are you using?

Hi @<1558986867771183104:profile|ShakyKangaroo32> , you can extract that using the API with events.debug_images - None

I suggest navigating to the UI and seeing what the UI sends when looking at debug samples with dev tools (F12)

Hi @<1718437476634202112:profile|SteepAnt48> , you certainly can! But you already have HPO capability built in:
None
None

Hi ExuberantParrot61 , if it's on the same machine you can use different configuration files. for example, when running the agent you can use --config-file <PATH_TO_USER_CONFIG_FILE>

You have a self hosted server, several agents that are running and two users developing on their own machines but the training is done elsewhere?

GrittyCormorant73 , K8s deployment will have easier time to spin up agent instances to run the tasks 🙂

Is this all happening when you're running locally? How many gpu's do you have/try to run on? Also, can you provide an example code snippet to try and run something basic to get a similar failure. I think I have a machine with multiple gpus that I can try playing on 🙂

Hi @<1523701295830011904:profile|CluelessFlamingo93> , I'm afraid there is no clear-cut way to migrate data from the community server to your own self hosted server since the databases aren't compatible.

One work around would be to pull all experiments information via API (The structure/logs/metrics) and then repopulate the new server using the API. I think it would be a bit cumbersome but it can be achieved.

I'm not sure it's possible to hide via the Task object. But I think such configurations should be saved as env variables or in clearml.conf

Can you please be more specific on the use case?

Hi @<1544853695869489152:profile|NonchalantOx99> , do you mean config parameters on the task itself?

I think in that case you should be using environment variables that are set on the machine and then access them via the task

Hi @<1524560082761682944:profile|MammothParrot39> , I think you need to run the pipeline at least once (at least the first step should start) for it to "catch" the configs. I suggest you run once with pipe.start_locally(run_pipeline_steps_locally=True)

That's strange indeed. What if you right click one of the pipeline executions and click on run?

Hi @<1595225620323766272:profile|VivaciousElk3> , you can set it with env variables - None

The ones that you're looking for are:
CLEARML_AGENT_SKIP_PIP_VENV_INSTALL
CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL

Hi @<1523701842515595264:profile|PleasantOwl46> , I'm afraid that such a capability doesn't really exist in ClearML. You could technically populate an experiment using the API.

I'm however curious - what is your use case for this?

Sounds about right. You can also create reports from the best experiments and then reference the models via the report as well

Hi ShallowCormorant89 ,

When does 1. happen? Can you add the full log?

Regarding 2, can you please elaborate? What is your pipeline doing and what sort of configurations would you want to add? On the pipeline controller level or steps?

Hi BattyDove56 , it looks like your elasticsearch container is restarting. Is this the issue still? Can you check the container logs to see why it's restarting? I think this is what might be causing the issue with ClearML server not raising properly

They do look identical, I think the same issue (If it's an issue) also affects https://clear.ml/docs/latest/docs/references/sdk/dataset/#list_added_files

Hi DistressedGoat23 , can you please elaborate a bit on what you're like to do?

DistressedGoat23 , how are you running this hyper parameter tuning? Ideally you need to have

` From clearml import Task

task = Task.init() `
In your running code, from that point onwards you should have tracking

And when you run it again under exactly the same circumstances it works fine?

Can you look in the UI if the execution parameters were logged?

Hi @<1554638166823014400:profile|ExuberantBat24> , you mean dynamic GPU allocation on the same machine?

Hi CrookedWalrus33 , I think this is what you're looking for:
https://github.com/allegroai/clearml-agent/blob/master/docs/clearml.conf#L78

My guess other agents are sitting on different machines, did you verify that the credentials are the same between the different clearml.conf files? Maybe @<1523701087100473344:profile|SuccessfulKoala55> might have an idea

I think this is an enterprise only feature

Hi @<1523701304709353472:profile|OddShrimp85> , you can do dir(object) to see what options you have. I think it would be something like <DS>.id

Hi @<1719162252994547712:profile|FloppyLeopard12> , not sure what you're trying to do, can you please elaborate?