clearml

clearml copied to clipboard

pytest.PytestUnraisableExceptionWarning : Error raised with program stop while testing the pipeline controller class using pytest.

3

Created two separate python file and while testing the file using pytest it is raising exception with error : PytestUnraisableExceptionWarning Details : test1.py : Created a pipeline controller to execute...

Kapil-23

task.set_packages do not recognize requirements

1

## Describe the bug When I try to set packages, it gives me this error: File "train.py", line 102, in train task.set_packages(packages=Path(root/'requirements.txt')) File"\Lib\site-packages\clearml\task.py", line 1595, in set_packages self._update_requirements(packages or "")...

andruum

Allow to set the `working_dir` with pipelines from decorators

1

## Proposal Summary Add a `working_dir` parameter to `PipelineDecorator.pipeline` and `PipelineDecorator.component` to be able to set the execution working_dir. ## Motivation Currently when defining a pipeline using `PipelineDecorator.component`, it's only...

materight

GPU Utilization for MIG devices

2

## Proposal Summary Currently I'm using MIG devices to spin up agents that serve different ClearML queues, however, when looking at the Workers CPU and GPU usage graph, I see...

uzmargomez

[Feature request | clearml-agent] allow to stop a specific agent from the command-line option by incorporating arguments to `--stop`

4

For now, the only way I know to kill a particular agent from ClearML API is through the arguments that define it (those that are entered when spinning it up)....

talajasi7

no module named "virtualenv" with execute_remotely

1

As title, what should I do to install some dependencies before running scripts? I use task.execute_remotely() with local git repo, and I get the error message from clearML server web...

AlgorithmicIntelligence

## Proposal Summary In order to be able to make flow heavy dataframes across several components, it could be beneficial to allow to set a compression level to the return...

MightyGoldenJA

Adding a non-clearml decorated function to PipelineController with add_function_step fails

2

## Describe the bug When adding a decorated function, a custom decorator, to a PipelineController with add_function_step, the created code for the pipeline step is wrongly indented and hence the...

jmw52

Using `target_project` with `TaskScheduler.add_task()` corrupts project?

6

## Describe the bug When I use the [`TaskScheduler.add_task()`](https://clear.ml/docs/latest/docs/references/sdk/scheduler#add_task) function to schedule a task, and I specify a `target_project`, I get an error from the running pipeline that it can't...

olucafont6

ClearMLCallbackHandler.on_chain_end and ClearMLCallbackHandler.on_chain_start raise exceptions of missing key

4

## Describe the bug In my langchain program, I constantly get errors with missing key `output` in `outputs` and `input` in `inputs`: ``` WARNING:langchain.callbacks.manager:Error in ClearMLCallbackHandler.on_chain_end callback: 'output' WARNING:langchain.callbacks.manager:Error in...

mdziezyc