Hello, I have another, hopefully minor question. What is the recommended way of checking if a code segment is running as a part of a Task or not? I tried che...

hi, i want to run a script remotely on my agent, but for it to work i need it to download to the agent the whole directory the script is in, is it possible?

Is it possible to add a callback for a pipeline from a step?

Hi, ClearML console leaks credentials passed in as Env Vars. The issue remains with clearml version==1.1.1.135 - 1.1.1 - 2.1.4 (As listed on the profile page...

Hi, we saw 2 more issues with images logging: 1. You need to use tf.summary.image and not summary_ops_v2.image 2. Image needs to be in range [0, 1] and not [...

How come I use logger.report_matplotlib_figure and get as a result a plotly figure??? O_O

Last reminder for our first community talk! We'll be discussing our latest feature - ClearML Apps! We'll also review the new ClearML Roadmap and have a Q&A S...

Hi, not sure if I'm doing something wrong or I found a bug. When I try to overwrite some parameters in a cloned task using get_parameters and set_parameters ...

Hi! I'm currently saving a dataframe with predictions inside the task. To do so, I save a dataframe as pickle file in /tmp/predictions_dataframe.pkl , and th...

Hi! I am trying to provide a custom output-uri for a task using clearml-task command. In our case it is a path to a data storage, which would be available to...

Question: has anyone done anything with Ray or RLLib, and ClearML? Would ClearML be able to integrate with those out of the box? https://medium.com/distribut...

Is it possible to launch a clearml-task from the command line to a remote machine while loading a local .env file as a configuration object?

When using PipelineControler is there a way to execute locally? Or must I use a queue?

Hey all. Question about the bucket config for clearml. Is it possible to have two buckets in the config? We have partners in EU and US, and due to data gover...

I want to use the PipelineController which assumes a task already been executed and as ready as template. I never executed the task before and it is a very l...

In PipelineController , we can set a flag disable_clone_task to re-use the scheduled task. Initially that task is in the draft mode, so this works on the fir...

I have to say I'm totally confused by the pipeline I want to execute the pipeline on my local computer. I followed https://github.com/allegroai/clearml/blob/...

Hi, we noted that using K8S glue, there are some situations where the task cannot be registered as error and will be stuck at pending. An example of one situ...

Hi, I want to pass environment variables from the host to the docker containers running my task. I managed to use extra_docker_shell_script: ["export SECRET=...

Any idea why I get this error in all my agents clearml_agent: ERROR: APIError: code 400/707: No queue is tagged as the default queue for this company

When viewing scalars is it possible to: Have a grid view (e.g. 3 plots per line instead of just one) Group the metrics differently (seems like if I log X/a a...

Hi, I want to install a local package using our package index, but I'm struggling with trying to make pip trust this host. If I want to install it in my venv...

Hello, is there any way to download all scalars, not only last metrics, by python interface. I am going to analyze my learning curves by myself.

Hi, Is there any reason why artifacts linked to a task are not removed when the task is removed from the experiment list?

hey guys, is there a ready script that can delete all models from S3 (or other storage) that are related to deleted or archived experiments?

It is possible to attach to an OutputModel an object closely related to it (as some product of data preprocessing that has been done specifically for that mo...

Hello I”m getting a ‘Fetch Experiments failed’ on the UI. How can one solve this

If I have a task and a dataset is being created in a task, how can I get a “link” that this Dataset is created in this task, similar to how Model has the tas...

Let's say that I specify the output_uri parameter in Task.init like this: task = Task.init( project_name="example_project", task_name="example_task", output_...

Hi! I am setting up ClearML server with web authentication. As far as I understand, users use logins and passwords specified in config/apiserver.conf to acce...