
Reputation
Badges 1
80 × Eureka!i’m not sure if i use the command correctly
i tried to add environment right before importing clearml, but it doesn’t work as expectedos.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 experimentsClearML 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 thisos.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?
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!
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