'CLEARML_CONFIG_FILE': '/home/ubuntu/clearml.conf'
I found that I can change Kernel and choose env
This looks more appropriate if the username itself is "ubuntu"
I'm being silly. You're actually directing it to the file itself to where it resides
'CLEARML_CONFIG_FILE': '/home/ubuntu/notebooks/clearml.conf'
os.environ['CLEARML_CONFIG_FILE']='~/clearml.conf'
Can I assume that you've installed the latest version of ClearML?
Are you sure that the file is on the server? Can you access it?
You must perform Task.init()
to have something reported 🙂
Successfully installed clearml-1.4.1 furl-2.1.3 orderedmultidict-1.0.1 pyjwt-2.1.0
OMG I tried different environment and started everything from scratch and it seems it works…
I can ssh for sure but I do not know how to check the environment 😳
I tried different references and nothing works. funny enough that clearml-init sees the file and reports “Configuration file already exists: /home/ubuntu/clearml.conf” but then the task.init does not see it
when I check os.environ I see'CLEARML_CONFIG_FILE': '/clearml.conf'})
Because now I have successfully import clearml and task but have an error with task initializingValueError: ClearML configuration could not be found (missing
~/clearml.confor Environment CLEARML_API_HOST) To get started with ClearML: setup your own
clearml-server, or create a free account at
I changed and environment and I think it helped to install clearml. But now I can not get how it can use this my file with conf.
Below is my code:
import os
os.environ[‘CLEARML_CONFIG_FILE’]=‘/clearml.conf’
and then I use import taks.
But if I do not use init, should I export somehow params before?
with such command: !export CLEARML_CONFIG_FILE=/root/clearml.conf