Reputation
Badges 1
662 × Eureka!I guess the thing that's missing from offline execution is being able to load an offline task without uploading it to the backend.
Or is that functionality provided by setting offline mode and then importing an offline task?
Last but not least - can I cancel the offline zip creation if I'm not interested in it π€
EDIT: I see not, guess one has to patch ZipFile
...
I dunno :man-shrugging: but Task.init is clearly incompatible with pytest and friends
I'll try it out, but I would not like to rewrite that code myself maintain it, that's my point π
Or are you suggesting I Task.import_offline_session
?
Seems like Task.create
is the correct use-case then, since again this is about testing flows using e.g. pytest, so the task is not the current process.
I've at least seen references in dataset.py
's code that seem to apply to offline mode (e.g. in Dataset.create
there is if output_uri and not Task._offline_mode:
, so someone did consider datasets in offline mode)
I think also the script path in the created task will cause some issues, but letβs seeβ¦
This is with:Task.set_offline_mode(True) task = Task.init(..., auto_connect_streams=False)
Or if it wasn't clear, that chunk of code is from clearml's dataset.py
I've been answering there as well π€
Thanks! To clarify, all the agent does is then spawn new nodes to cover the tasks?
None, they're unusable for us.
Any updates @<1523701087100473344:profile|SuccessfulKoala55> ? π«£
Honestly I wouldn't mind building the image myself, but the glue-k8s setup is missing some documentation so I'm not sure how to proceed
No task, no dataset, just an empty container with no reference to the task it's attached.
It seems to me that it should not move the task if use_current_task=True
?
I believe that happens natively thanks to pyhocon? No idea why it fails on mac
I guess I'll have to rerun the experiment without tags for this?
Thanks for the reply @<1523701827080556544:profile|JuicyFox94> ! I'll debug more and let you know
Running a self-hosted server indeed. It's part of a code that simply adds or uploads an artifact π€
So the pipeline runs successfully, I can find all the different tasks, but I cannot see them in the Pipelines tabβ¦
Happens with the latest version indeed.
I canβt share our code, but the gist of it is:
pipe = PipelineController(name=..., project=..., version=...)
pipe.add_function_step(...) # Many calls
pipe.set_default_execution_queue(...)
pipe.start(queue=..., wait=True)
I believe that a Pipeline should have the system tags ( pipeline
, maybe hidden
), even if it created in a running Task
.
I can navigate through the projects, but selecting one task in one project, then navigating to another project and selecting a different task -> there is no suggestion to compare the tasks.
In the projects page if I show all - I just see the projects. If I search for a task of similar name, I get results, but I can't compare them via the UI.
The only way I managed so far was to create a pseudo-comparison between unrelated tasks in the same project, then remove one task from comparion, and u...
Hey AgitatedDove14 π
Finally managed; you keep saying "all projects" but you meant the "All Experiments" project instead. That's a good start π Thanks!
Couple of thoughts from this experience:
Could we add a comparison feature directly from the search results (Dashboard view -> search -> highlight some experiments for comparison)? Could we add a filter on the project name in the "All Experiments" project? Could we add the project for each of the search results? (see above pictur...
Unfortunately I can't take a photo of not being able to compare tasks by navigating around the WebUI...
Does that clarify the issue CostlyOstrich36 ?
That's exactly what I meant AgitatedDove14 π It's just that to access that comparison page, you have to make a comparison first. It would be handy to have a link (in the side bar?) to an empty comparison
I can't seem to manage the first way around. If I select tasks in different projects, I don't get the bottom bar offering to compare between them
For example, can't interact with these two tasks from this view (got here from searching in the dashboard view; they're in different projects):
Either, honestly, would be great. I meant even just a link to a blank comparison and one can then add the experiments from that view