Hi ExasperatedCrocodile76 , I think this is the quickest and easiest solution you're looking for:
https://clear.ml/docs/latest/docs/references/sdk/task/#taskforce_requirements_env_freeze
Just make sure to call it before Task.init() as documentation states

OddShrimp85 , Can you see if the models are connected to the experiment in the UI and if they are, what their uri is? What was your use case?

Is it failing when running on a worker?

I'm not sure I understand. Can you give a specific example of what you have VS what you'd like it to be?

Can you please open a github issue so the issue can be followed?

Hi @<1523704157695905792:profile|VivaciousBadger56> , you can see an example of usage of Task.connect() here:
None

Hi @<1593413673383104512:profile|MiniatureDragonfly17> , no. The assumption is that serving runs on a dedicated machine. Of course you can edit the docker compose to use different ports

Hi PunyWoodpecker71 , I guess it means the task will never finish either

Shouldn't be 🙂
Did you notice any difference?

Hi FierceRabbit20 , I don't think there is such an option out of the box but you can simply add it to your startup of the machine or create a Cron job

Hi @<1557174909573009408:profile|LargeOctopus32> , I suggest you through the introduction videos on ClearML's youtube channel
None

DrabSwan66 Hi!

What version are you trying to install? If the machine in question has an issue installing opencv then the agent will most likely fail as well.

Looks like you're trying to do fetch something and then do something to it, that's why there's a TypeError regarding a NoneType. Do you know where in the code the traceback occurs?

JitteryCoyote63 , projects.get_all_ex is fired when UI is loaded/pages are navigated to.
Regarding the self hosted version channel, usually these things are discussed and revealed in community talks that are done once in a while. Considering that the community server was recently updated I would give an educated guess of a week or two until ClearML self hosted version is released 🙂

But I believe you can use your user/group if you prefer 🙂

JitteryCoyote63 , heya, yes it is :)
You can save the entire folder as an artifact.

https://github.com/allegroai/clearml/blob/master/examples/reporting/artifacts.py

What's the use case?

Hi OutrageousSheep60 ,

I'm not sure I've seen such an example. But you can definately wrap torch dataset in clearml datasets. I am over simplifying but in the end think of the clearml dataset is like a large zip of all your files.

If you upgrade to the latest versions and this issue still occurs it will be possible to debug but your current version is very old

Hi UpsetSheep55 ,

Permissions feature is indeed exists only in the enterprise version. There are no examples for this since this an enterprise only feature.

ClearML should be backwards compatible - any combination will work. However it's always suggested to use the latest versions 🙂

And you get the error only when there are no credentials or is this unconnected?

Doesn't look like anything critical 🙂

Hi UnevenDolphin73 , is there a specific setting you're looking for?

Hi DrabCockroach54 , what version of ClearML are you using? What OS is this running on?

Can you please open a GitHub issue to follow up on this issue?

Hi EcstaticBaldeagle77 ,
I'm not sure I follow. Are you using the self hosted server - and you'd like to move data from one self hosted server to another?

Hi @<1769534182561681408:profile|ReassuredFrog10> , do you have a GPU available? Maybe try the other docker compose without Triton as that one is specifically built for GPU inference.

You have the open source repository of the documents - None

I think you could generate a pdf from that with some code.