How would that be reflected on the plot?

Additional info:
-Public URL uses HTTPS, internal traffic doesn't.
-clearml.storage fails while trying to fetch None ...

Meaning it just replaced the internal IP with the URL at some point for some reason, it doesn't exist in that form anywhere in any configs (http and public URL).

clearml-1.13.1

Task.add_requirements("requirements.txt")
task = Task.init(project_name="My project", task_name="My task")
task.execute_remotely(queue_name="default")
...

So after publishing a task (right click/Publish from WebUI), one of the models got their id changed to __DELETED__4be00...
The other one (last_model on the screenshot below) is all good and didn't get deleted in this way.
"best_model" exists on the disk and I can access it by taking last_model's URL and just changing the file name, but I cannot normally access it via id (which has now changed to __DELETED__4be00...). Any ideas why this might have happened?
![image](https://clearml-web-assets....

Oh, I misunderstood then docs/examples, sorry. I'm using pytorch-ignite.

Thanks for the tip!

Got it, thanks a lot!

Having a bit of trouble with this one (sorry for possibly dumb questions).
Are there any docs on how to add certs to the docker image? I see this ( None ) which is where letsencrypt points me to, but I'm not sure what's the proper way to do this on the webapp docker (I'd assume there's a non-hacky way to do it as others are using the same setup I'm trying to make work I guess)

I'll try to reproduce it and will get back at you. The HPO task (parent of this task) was deleted indeed but that shouldn't matter? One of the models was deleted but the other one wasn't.

In the Task info tab there is no GPU and pytorch doesn't see the cuda device.
I'm not sure how to debug it, that would be my first question. So I should first check if docker is executed with --gpus? I'll pay attention to this next time this happens, thanks.

Kind ping on this thread, thanks! 🙂

Got it, thanks!

"Executing: ['docker', 'run', '-t', '--gpus', '"device=0"'" - so the container is executed with --gpus.
However, torch.cuda.is_available() returns False.

Failed to initialize NVML: Unknown Error

Got it, thanks folks!

Weird. When I spawn agent with sudo I get this behaviour. Without sudo everything works fine

    model_checkpoint = ModelCheckpoint(
        "checkpoint",
        n_saved=2,
        filename_prefix="best",
        score_function=score_function,
        score_name="accuracy",
        global_step_transform=global_step_from_engine(trainer),
    )

    # Save the model after every epoch of val_evaluator is completed
    val_evaluator.add_event_handler(
        Events.COMPLETED, model_checkpoint, {"model": model}
    )

The issue was .ssh wasn't propagated so the git repository couldn't be cloned.

Ooooh, I didn't notice that field is editable. Thanks!
So I should use add_requirements before Task.init and delete the list from webUI when needed?

I've ran it before Task.init

I've tried that one, but it behaves the same :/

Makes sense, thanks!

@<1523701087100473344:profile|SuccessfulKoala55> kind reminder not to miss this when you catch time, thanks!

I just added the secrets/keys to docker-compose.yml and restarted everything but no change.

To make sure I understand, I need to setup a domain with a cert and it should work, no additional ClearML config is required?

One more related question (I hope there's a similar solution), when I log images, they appear in the UI with http://<my-ip> so they are inaccessible (they should be translated to None . Is there any path_substitution variant for this scenario in the config? I can't seem to find it in the docs. Thanks!

I hacked around the solution by setting api.files_server for the agent to the public URL, but ideally I'd avoid going through reverse-proxy if there's some path_substitution equivalent for this. Thanks

@<1523701087100473344:profile|SuccessfulKoala55> Kind reminder again, thanks and sorry!

Not ClearML employee (just a recent user), but maybe this will help? None

Got it, thanks!

No worries, sorry for pinging, was just making sure you (or anyone else who might help) doesn't miss it 🙂
I use Task.add_requirements("requirements.txt") right before the Task.init.
In main, I parse arguments command-line, add_requirements, initialize Task and call execute_remotely. After that it's all pretty much the usual workflow. Initialize the model, setup dataloaders, optimizer and run the training. I'm using pytorch-ignite and have model checkpoint made on validation evaluator COMPL...