TrickyFox41 are you saying that if you add Task.init inthe code it works, but when you are calling "clearml-task" it does not work? (in both cases editing the Args/overrides ?

I used an env variable to avoid creating and endless loop of init/enqueue (using an argument like clearml.queue that would be captured and forwarded to the agent)

I tried like that clearml-task --script train.py --args overrides="log.clearml=True train.epochs=200 clearml.save=True" --project mornag-plain-dense --name mornag-plain-dense-training --queue tesla-t4 --skip-task-init

Hmm that should have worked ...
I'm assuming the Task itself is running on a remote agent, correct ?
Can you see the changes in the OmegaConf section ?
what happens when you pass
--args overrides="['dataset.path=abcd']"

same result. in print(cfg) there are no overrides

--args param.ovveride=value does not work with clearml-task

I see TrickyFox41 try the following:
--args overrides="param=value"Notice this will change the Args/overrides argument that will be parsed by hydra to override it's params

So I think there are two bugs here?
--args overrides="key=value" does not work request: add --hydra to override hydra arguments (and if this is added the first one is not needed)Is that correct?

Hmm can you try:
--args overrides="['log.clearml=True','train.epochs=200','clearml.save=True']"

Hi TrickyFox41

Hey since Hydra does not work with

clearml-task

I should shouldn't it? what does not work ?

What do you have under Hydra section? and OmegaConf section

By the way, since if i create the task locally, reset it and enqueue it, it works. This is the workaround that i'm using right now

It only works with argparse

AgitatedDove14 Yes exactly!

No luck with --args overrides="['dataset.path=abcd']"

i printed cfg in the script and the config has not been overwritten 😢

Yes the task is running on a remote agent with the --docker flag
this is the config on the machine(s) running the agent
`
agent {
venvs_cache: {
max_entries: 50
free_space_threshold_gb: -1
path: ~/.clearml/venvs-cache
}
extra_docker_arguments: [
"--network", "host",
"-v", "/home/ubuntu/.ssh:/root/.ssh:ro",
"-v", "/home/ubuntu/.cache:/root/.cache",
]

docker_internal_mounts {
    sdk_cache: "/clearml_agent_cache"
    apt_cache: "/var/cache/apt/archives"
    ssh_folder: "~/.ssh"
    ssh_ro_folder: "/.ssh"
    pip_cache: "/root/.pip-cache"
    poetry_cache: "/root/.pypoetry-cache"
    vcs_cache: "/root/.clearml/vcs-cache"
    venv_build: "~/.clearml/venvs-builds"
    pip_download: "/root/.clearml/pip-download-cache"
}

}

sdk {
aws {
s3 {
host: "...:443"
key: "..."
secret: "..."
secure: true
}
}
}

api {
web_server: https://...
api_server: https://...
file_server: https://...

credentials {

}

} `

This is what it looks like from the web app

Related GitHub issue https://github.com/allegroai/clearml/issues/847