Hi @<1673501397007470592:profile|RelievedDuck3> , no you don't. The basics can be run with a docker compose 🙂

Are you running a self hosted server?

And the pipeline runs with agents or locally?

Hmmmm I think you would need to change some configurations in the docker-compose to use https

Hi @<1543766544847212544:profile|SorePelican79> , can you please elaborate a bit?

You can view all projects and search there 🙂

I played a bit with it and got to the value. OutrageousSheep60 , please tell me if this helps you 🙂

` >>> task.set_user_properties(x=5)
True

y=task.get_user_properties()
y
{'x': {'section': 'properties', 'name': 'x', 'value': '5'}}
y["x"]["value"]
'5' `

Hi @<1706116294329241600:profile|MinuteMouse44> , is there any worker listening to the queue?

Hi @<1760474471606521856:profile|UptightMoth89> , what if you just run the pipeline without run locally and then enqueue it (assuming you have no uncommitted changes)

As long as you're working with a local git I'm not sure it will log the branch/working_dir/repo

However, you can probably circumvent this maybe with configuration objects

There are constant upgrades in the webUI, so there is always a good chance it can fix it

Hi @<1654294834359308288:profile|DistressedCentipede23> , do you mean completely circumventing that basic function of the agent and simply pointing it to a specific script/env?

Hi @<1523701624541810688:profile|RotundHedgehog76> , how are you running the agent? What is the command you're using?

Makes perfect sense, just careful not to run out of memory or it'll crash everything 😛

Hi @<1556450111259676672:profile|PlainSeaurchin97> , I think what you're looking for is the output_uri parameter in Task.init()

SmugTurtle78 , regarding the CPU only mode - How are you running. Are you using the application in PRO version or are you running through one of the examples?

You can also set just specific commit/branch in the component object - None

Hi ShinyRabbit94 ,
Do you get any error when you attempt to delete the Tasks?

Hi RoundMosquito25 , if you already have an instance of agent running on a queue and you run clearml-agent execute --id <task_id> this will create another instance of the agent and will run the experiment regardless of what is happening with the other agent

Do you mean reordering them with a drag and drop functionality or options for different orderings?

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?

Try adding the following to your Task.init()

task = Task.init(project_name='<PROJECT_NAME>', task_name='<TASK_NAME>', output_uri=True)
Another option that will solve the issue as well is editing your ~/clearml.conf and changing the following section sdk.development.default_output_uri to point to your fileserver/hosting

Hi UnevenDolphin73 , what are you trying to do when you get this? Are you running a self hosted server?

If you mean to fetch the notebook via code you can see this example here:
None

What do you mean exactly by run it as notebook? Do you mean you want an interactive session to work on a jupyter notebook?

Hi CooperativeOtter46 ,

I think the best way would be to use the API (You can use the SDK but I don't think it is so easy to filter times)
Use https://clear.ml/docs/latest/docs/references/api/tasks#post-tasksget_all api call to to get all the tasks according to the time frame & filtering you want and then sum up the run time from all experiments returned 🙂

Also, in the original experiment, what pytorch version is detected?

Hi @<1590514584836378624:profile|AmiableSeaturtle81> , please try to keep your messages to a single thread and not spamming the main channel on the same topic 🙂

Hi StormyBluewhale12 , HPO pushes the tasks into a queue. You need an agent connected to the queue that will start executing tasks.

Heya ReassuredTiger98 ,

Is there anything specific you're looking for?

SkinnyPanda43 , I think so yes 🙂