here is the key/secret

sdk {
    aws {
        s3 {
            region: "ap-northeast-2"
            use_credentials_chain: false
            extra_args: {}
            credentials: [
                {
                    bucket: "

"
                    key: "S3_KEY"
                    secret: "S3_SECRET"
                }
            ]
        }
    }
}

Thank you for your advice!
I will change the s3 bucket that does not have a dot and try again.

this is my client side’s clearml.conf file
I think it is almost similar with agent’s clearml.conf file

@<1523701087100473344:profile|SuccessfulKoala55> I changed s3 bucket name None , but still has same error above.

here is a full clearml.conf file

after restarting the docker-compose, then another error appeared

---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
Cell In[6], line 1
----> 1 StorageManager.list("

")

File ~/.clearml/venvs-builds.2/3.8/lib/python3.8/site-packages/clearml/storage/manager.py:452, in StorageManager.list(cls, remote_url, return_full_path, with_metadata)
    430 @classmethod
    431 def list(cls...

Thank you for your advice @<1523701087100473344:profile|SuccessfulKoala55>
I am really appreciate it.

And this is the task configuration info

here is the full log of the failed task

And this is the HPO’s configuration info

Yes, in the train.py, I put the task.init

Thank you for giving me the advice @<1523701087100473344:profile|SuccessfulKoala55> @<1523701205467926528:profile|AgitatedDove14> !!

Thank you for giving me the advice.

To answer your question, here is my workflow.

First, I create the task by running the below code

python3 train.py config object_detection.yaml

And in the same docker image, I run the below command to executing an agent

clearml-agent daemon --queue default --forground

After that, use this task id created above, I run the code I shared clearml_hyper.py

So I think argparser arguments are injected in the task itself before HPO

and this is the Nanodet modified train code