FancyWhale93 , you need to make the same configuration changes to ~/clearml.conf
file on the machine running clearml-agent
that is running the pipeline steps
Thanks for the reply CostlyOstrich36
The local machine & agent machine has the same clearml.conf
. As I sad, storing something via StorageManager
has no problem on both side.
But PipelineDecorator.upload_model(model_name, model_path)
doesn't work on both side.
Hi FancyWhale93 , in your clear.conf configure default output uri, you can specify the file server as default, or any object storage:
https://github.com/allegroai/clearml-agent/blob/9054ea37c2ef9152f8eca18ee4173893784c5f95/docs/clearml.conf#L409
Great! setting the default_output_uri
worked!
Thank you.