When we run our code and it communicate with clearml server, is there some way we can log that api request? Like what endpoint is it and what payload it send...

Hello folks. We're a small team currently considering adopting ClearML for experiment tracking. I was wondering if I start with the hosted service and decide...

I'm new to using datasets, if my git project root is myProject and I expect file.json to be at the root level, how do I accomplish this?

If I have an AWS key/secret for an IAM user, what is the best way to pass in these credentials so the task docker container has credentials generated for usa...

Hey everyone ! Seems like i can not run my task remotely <- is something missing in the config (agent) , should i specify all the versions? I attach full log

Hey all, I'm testing the usage of SETUP SHELL SCRIPT in the experiment window. I added a simple command but did not see it in the console. The task did execu...

Hello everyone ! I would like to https://github.com/allegroai/clearml/blob/master/examples/pipeline/pipeline_from_tasks.py from https://github.com/allegroai/...

Are the docs broken? https://clear.ml/docs/latest/docs/release_notes/ver_1_7 gives me this

Hi everyone, Suddenly since 1 hour ago I cannot show the graphs in 'scalars' section (under results). When pressing the scalars button, clearml shows a 'wait...

Is the https://azuremarketplace.microsoft.com/en-us/marketplace/apps/apps-4-rent.clearml-on-centos8 maintained? I can't see anywhere a log of releases or som...

Hi all, where does the installed packages list populate from in the task viewer?

Is there any additional configuration needed for PYTHONPATH to be setup properly in the clearml agent? I'm getting python import errors from the root directo...

If I leave WORKING DIRECTORY empty in the experiment configuration (in the UI), will that use the git project root by default?

Does anyone know if it is possible to add a plot from a stylized pandas data frame? I can easily log a pandas data frame with logger.report_table but can I l...

Hello everyone! i quite confuse about update data using sdk without download it. first i create dataset and upload. dataset_path = 'dataset/batch_1' dataset ...

Assuming I want to report to the same task from two different processes (running on different machines) - how would I do that? So I have some function foo th...

Hello everyone ! My clearml-agent was working fine. After computer restart i got this error : clearml_agent: ERROR: create..Validator.init() got an unexpecte...

Hello all, When I create a pipeline (using pipeline decorator) and call the function executing_pipeline(). It create a pipeline in the "Pipeline" menu, with ...

Hi! I’m running an experiment as follows: create argparser -> args Task.init (default args) call train(args)Then I clone it and change the parameters in the ...

Hi, I am new to ClearML but already really impressed. I have what seems like a dumb question but asking it anyways... I have a matplotlib figure which is aut...

@<1523701205467926528:profile|AgitatedDove14> @<1523701087100473344:profile|SuccessfulKoala55> Hi, I have a question when using pipeline. I use pipeline with...

When I try to create experiment in the UI all I see is this dialogue

How do I create an experiment where I can set the github repository/branch name/script path like this example shows?

Hello guys! I have a little question: My metrics quota has been reached, and I cannot use the platform anymore.. I have already removed almost all projects s...

Hey, how can I download report_table objects back into python as a dataframe? I'm sure I've done this before but have no idea how I did it.

Hi, Quickhelp with pipelines: I am loading a model during a state of it and them passing this model (torch.nn.Module object) as input argument to a pipeline ...

Hello! When I use the TriggerScheduler class with the add_task_trigger function configured to watch for trigger_on_status=['published'] and a specific trigge...

Hello! When I squash multiple datasets (e.g. Dataset.squash(dataset_name="new_ds", dataset_ids=[id1, id2, id3]) , as far as I can see the newly created datas...

I am trying to run the urbandsounds8k example, but when I run "preprocessing" I get the error in the line self.metadata = Task.get_task(task_id=self.original...

Given I want to run a task in a pipeline using a base task id. One of my steps just finds the latest results to use. I want the task to output the id of the ...