i’m not sure if i use the command correctly

not actually

i tried to add environment right before importing clearml, but it doesn’t work as expected
os.environ['CLEARML_CONFIG_FILE'] = str(Path.home()/f"clearml-{socket.getfqdn()}.conf") from clearml import Task Task.init(project_name="Alfred", task_name="finalized", auto_connect_frameworks={'pytorch': False})

i only want to use clearml agent to change the conf file

but basically i create it by init and rename it

Yes, when i put the task init into the spawn function, it can run without error, but it seems that each of the child process has their own experiments
ClearML Task: created new task id=54ce0761934c42dbacb02a5c059314da ClearML Task: created new task id=fe66f8ec29a1476c8e6176989a4c67e9 ClearML results page: ClearML results page: `
ClearML Task: overwriting (reusing) task id=de46ccdfb6c047f689db6e50e6fb8291
ClearML Task: created new task id=91f891a272364713a4c3019d0afa058e
ClearML re...

what i want to do is to init one task and multiple works can log with this one task in parallel. TimelyPenguin76

I think this is not related to pytorch, because it shows the same problem with mp spawn

when i run this one clearml-agent --config-file ~/clearml-iris.conf it output the help info

i’m setting the environment variable in the python script like this
os.environ['CLEARML_CONFIG_FILE'] = str(Path.home()/f"clearml-{socket.getfqdn()}.conf")then task.init

unfortunately, no, when i try to click the link, there is nothing there https://demoapp.demo.clear.ml/projects/0d49bffcdaa441c2aa3224054737d0bd/experiments/26dd46ec11fd4f95ba522955820a8444/output/log

then i run the second one, it basically outputs the same as just init

so basically, the spawn will run a function in several separate processes, so i followed the link you gave above and put task.init into that function.
i guess in this way, there will be multiple task.init running.

i was trying to copy the content of that file

i just tried, but is still open result on public server.

Thank you! another question this method seems need to get the result one by one on the fly. Because i have lots of complete experiments, is there a way that i can pull all scalars at once? Or, can i get experiments list and pull the data?

yes

I’ll try it tomorrow and let you know if there is anything wrong

which client should I import for Client.queues ?

But it seems buggy

Just tried the code

I tried from clearml.backend_api.session import client no luck

Apiclient will report

Oh this is one line missing on the above code

Sure I'm right here with you

works fine awesome!

On sec

Or can I enable agent in this kind of local mode?

Guess I’ll need to implement job schedule myself

So is there any tutorial on this topic