Hi here. Is it possible to run clearml-agent with pre-ready pip virtual environment and avoid packages installation from requirments.txt/INSTALLED PACKAGES ?

Different question. How can I pass PYTHONPATH env variable to a task, run by agent (so python can find classes inside m subdirectories)?

One more strange behaviour of clearml-agent. Two tasks with the same code (except usage of task.execute_remotely()) generate different structure of output ar...

Hi. I have a task executed on clearml-agent, configured with agent.package_manager.force_repo_requirements_txt = true . But requirements,txt is not taken int...

Hi all. Is it possible to configure agent NOT to install requirements.txt/INSTALLED PACKAGES and to use some pre-ready environment (for example, virtualenv c...

Task.init() takes by default output_uri from clearml.conf configuration file (S3 bucket in my case). But underlined task created with Dataset.create() ignore...

Hi clearml. I'm trying to look at Datasets functionality (with the help of https://github.com/allegroai/clearml/blob/master/docs/datasets.md and https://clea...

Hi all. Is there possibility to download all the artifacts from some task?

Hi all. If I understand right, Dataset and any other task is Aborted after some time of inactivity. Can I configure it on some level (ideally on task level)?...

Is there some support of multi-machine training on ClearML level?