Hi @<1603560525352931328:profile|BeefyOwl35> , The agent uses it's own entry point, so yes you do need to specify it even if it's in the dockerfile π
Hi @<1715900788393381888:profile|BitingSpider17> , you need to set it in the environment where you are running the agent. Basically export it as an env variable and then run the agent
Hi π
When viewing the experiments tables you have a tab called 'models'. If the artifact was saved as a model, then you can find it there π
RoughTiger69 , you can also use Task.add_requirements Β for a specific package through the script
Example: Task.add_requirements('tensorflow', '2.4.0') Example: Task.add_requirements('tensorflow', '>=2.4') Example: Task.add_requirements('tensorflow') -> use the installed tensorflow version Example: Task.add_requirements('tensorflow', '') -> no version limit
REST API is language agnostic, you can just implement the API calls in any language you like. Same as you can do with python unrelated to the APIclient module in the SDK
Hi @<1546303277010784256:profile|LivelyBadger26> I'm afraid that in the free version everyone is an admin. In the scale & Enterprise licenses you have full role based access controls on all elements in the system (from experiments to which workers can be provisioned to whom)
RoughTiger69 Hi!
Regarding your questions:
You can use the following: Task.force_requirements_env_freeze(requirements_file='repo/some_folder/requirements.txt') before your task=Task.init(...) You can configure sdk.development.detect_with_pip_freeze=true in your ~/clearml.conf file for full env detection from the environment you're running from
@<1539417873305309184:profile|DangerousMole43> , I think for this specific ability you would need to re-write your pipeline code with pipelines from decorators
Hi @<1829328217773707264:profile|DiminutiveButterfly84> , how are you building the pipeline? Is it from tasks or from decorators?
can you add full log of the pipeline + step? So you mean its not cloning the utils properly or doesn't import them properly?
Hi @<1562610703553007616:profile|CloudyCat50> , can you provide some code examples?
GiganticTurtle0 , then I'd guess that's the task that would be returned π
Did you try?
Hi EnormousCormorant39 , how did it fail?
Hi @<1578555761724755968:profile|GrievingKoala83> , I'm afraid that's an Enterprise/Scale only feature
Hi @<1829328217773707264:profile|DiminutiveButterfly84> , Is the code part of some repository or its just a folder with some script files?
REMOTE MACHINE:
- git ssh key is located at ~/.ssh/id_rsa
Is this also mounted into the docker itself?
YummyLion54 , Hi
What versions of clearml/clearml-agent are you using?
When running it without the agent, do the hydra configurations show up properly in the UI?
Hi OddShrimp85 , you sure can! You can use the API. This one is useful for getting data about specific tasks. I think you'll have to sift through the response to find what you need π
EnormousWorm79 , Hi π
What do you mean by dependency structure?
Hi @<1524922424720625664:profile|TartLeopard58> , you mean the side bar on the left with projects/datasets/etc... ?
TrickySheep9 Hi! Thanks for the heads up, we'll take a look π
Hi @<1759749707573235712:profile|PungentMouse21> , you should be able to access machine logs from the autoscaler, this should give you a place to search
For example--packages "tqdm>=2.1" "scikit-learn"
Hi @<1674226153906245632:profile|PreciousCoral74> , you certainly can, just use the Logger module π
None
Can you elaborate on this? You mean package dependencies when experiments are executing via agent?
Hi AdventurousButterfly15 ,
When running code locally, how are the installed packages detected? Does it detect your entire venv or does it detect only the packages that were used?
When you look at the original task that appears in the UI, what are the requirements shown in the 'execution' tab?
Hi @<1600661428556009472:profile|HighCoyote66> , I'm not sure I understand, can you please elaborate on this?
- Upload image from device via the UI buttonWhat do you mean by device?
Hi FierceHamster54 , you have docker_args in https://clear.ml/docs/latest/docs/references/sdk/automation_controller_pipelinecontroller#pipelinedecoratorcomponent
Hi @<1523701491863392256:profile|VastShells9> , the GCP autoscaler is not available in the open source I'm afraid. Only in PRO licenses and up