Hi guys, I'm setting up a bunch of machines as clearml agents and have run into an issue with caching. We are using Poetry for python dependency management, ...

Hi guys, I have set up the clearml-server on aws ec2 instance using the released clearml AMI in EC2 instance (there is a 24/7 running EC2 instance for runnin...

Hi guys ! I'm adding a requirements.txt to a task before initializing it. Is it the way it is supposed to look like on the UI ? Shouldn't I see directly the ...

Hi guys, can we do dropdown this on configuration?

Could some help me figure out why figures have disappeared from a report, please? My colleague made this report 6 weeks ago ago None , but all the graphs hav...

Hey, I have a question regarding “logger.report_table”, it’s seems like after the table is drawn in the UI I cannot change the column size and more annoying ...

Is there any way to change the x-axis on the charts for scalars, to say e.g. "epochs" instead of "iterations"? Or is that hardcoded?

Is there a way to upload an artifact I forgot to upload during the task duration to that task after it is already complete?

Hi guys. Say that we train a model with 10 epoch, and suddenly interruption occur on epoch 5. How can we continue the by using ClearML?

Hi all, I have a big amount of experiment in a single project, there is a way to download the experiments as a custom table? to let's say to a .csv file? Tha...

Hi guys! What is the best way to access artifacts from other step of the pipeline? I have step one returning dataframe and step two takes it as an input but ...

Hi, I'm looking into deploying ClearML with AWS Opensearch as managed Elasticsearch backend. I'm however not finding any documentation on whether ClearML sup...

Hi everyone, I’m new to ClearML, and our team has started investigating clearML vs MLflow. We’d like to try out the K8s setup using the helm charts, but afte...

Hi, Is this a well known issue, that running a task in a virtual environment, messes up the reproducibility feature ?

Hi all, in the AWS scheduler, what’s the difference between: - Regular Instance Rollback Timeout - Spot Instance Blackout Periodplease?

hey while trying to upload a (~14 gb) dataset, we get the following error:

Hello all 🙂 Does anyone knows how to select all models in "model archive" and delete them ? The button is grey for me... However, I can select all and archi...

is there a support channel?

I'm getting this error across different machines ValueError: Failed uploading object /crunch-pipeline/.pipelines/crunch-pl/crunch-pl %23950.c4c39b676e5643bca...

Hi! I'm looking to setup a periodic backup of ClearML self-hosted server which would ideally happen without shuting the server down. I'm guessing just copyin...

Hello everybody, I'm seeking for a clarification on the 'Metrics' quote for the SAAS platform. My workspace only uses self S3 storage as the File Server to s...

Hey guys, first of all ClearML rocks! I love it 🤩 Trying to log a figure, I got : upstream connect error or disconnect/reset before headers. reset reason: c...

Hello everybody, I am new to Clearml and could use your help with something. I am generating data and setting the metadata using ds.set_metadata(metadata='pa...

None

Hey there, I am a new user of clearml and really enjoying it so far! I noticed that my model checkpoints are saved after each epoch. Instead I would like to ...

Hello all, I’m an ML engineer looking to transition our company to a new MLops system. Many of our projects are currently built around Hydra and I’m attempti...

i am using hydra. When running locally the code takes into consideration any hydra override arguments i pass in the command line. but when i run the same tas...

Is there a way to pass environment variables when a job is executed over a queue? (either on the agent side or on the task.init side). I need this fix for po...

I have a Kubernetes setup of ClearML 1.7.2 done with the official Helm charts. When i upgrade to the latest version of ClearML with the newest Helm charts, w...

Hey friends! At our startup we are a big fan of ClearML and use it daily to run trainings and other tasks. Right now we are re-evaluating our data platform a...