CostlyOstrich36

0 Questions, 4213 Answers

Active since 10 January 2023

Last activity 2 years ago

Reputation

Answers 4213

0 Hey There, Is It Possible For A Clearml Pipeline Step To Log A Folder Instead Of Numpy/Pickle Objects? Looking At The Docs,

So in which scenario do you want to keep those folders as artifacts and where would you like to store them?

3 years ago

0 Good Day, We Have Been Using Clearml For Project Monitoring And Task Management. Recently, We Decided To Try The Google Cloud Platform Autoscaler To Automate Our Existing Gcp Vm Creation Pipelines, Task Queuing, And Processing. However, We'Ve Encountere

Hi @<1826791494376230912:profile|CornyLobster42> , it looks like there might be an issue with the image. Have you tried other images? From what I see here - None

Many people are talking about various issues with various solutions.

What if you try with this image? projects/ml-images/global/images/c0-deeplearning-common-cu121-v20231209-debian-11

6 months ago

0 Hi, I Have A Remote Task That I Want To Run On Some Of The Gpus Of The Remote Machine. I Set The Agent In The Remote Machine On Two Of The Gpus. How Do I Query The Agent What Are His Gpu So That My Code Will Use Only Them

Sorry, GPUs 🙂

2 years ago

0 Hello, I Want To Use Custom Packages In My Code.It Works Fine When I Run It Locally.But When I Clone The Experiments And Run On Agent It Shows Package Not Found Error In Agent. Can Anyone Help Me Out How To Use Custom Packages In Clearml? Which Parameter

Hi ShaggySquirrel23 , is this package inside some artifactory?

The agent needs access to the package while running, you need to have it accessible somehow on a remote machine as well. What is your setup?

2 years ago

0 Hi! Any Idea Why Clearml Fails To Detect Iteration Reporting?

GrievingTurkey78 , let me take a look into it 🙂

4 years ago

0 Hi All, I'M Trying To Create Pipeline With Decorator. When I Run The Code To Push The Pipeline To Clearml Server, It Seems The Pipeline Is Pending In "Services" Queue. But My Worker Is Listening To "Default" Queue, So The Pipeline Is Not Executed. I Read

Hi PunyWoodpecker71 ,

It's best to run the pipeline controller in the services queue because the assumption is that the controller doesn't require much compute power as opposed to steps that can be resource exhausting (depends on pipeline of course)

3 years ago

0 Hello Everyone! I Launched An Ec2 Instance That'S Free Tier (

CluelessElephant89 , Hi!

It looks like there is a problem with the API server. Can you please look for the docker logs and see what errors that it prints and paste here 🙂

4 years ago

0 Any Idea What I'Ve Missed Here? Thanks

BattyDove56 , the warning doesn't seem related. As I mentioned before, you need to check the elastic logs to see what's the issue. post them here so we can look together 🙂

3 years ago

0 Have You Guys Forgot To Add Uv Caching? Im Trying On 1.9.3Rc0 And Uv Always Redownloads The Packages, Doesnt Use Cache. Or Is It My Skill Issue? All I Added: Package_Manager: {

Hi @<1590514584836378624:profile|AmiableSeaturtle81> , did you have a chance to try out the solution suggested in GitHub and play with it a bit?

10 months ago

0 All Of Our Users Suddenly Cant Log In. We Get 401, Existing Logged Users Are Fine, But Once Logged Out Cant Log Back In. Here Is What Webserver Says In Logs:

Always good to use the latest 🙂

one year ago

0 Hi All, I'M Trying To Run Tasks In The Autoscaler Using My Docker Image That Sits In Aws Ecr, And I'M Having Authentication Credentials Issues. I Want To Clone An Existing Task And Run It Using The Autoscaler. I Created A Docker Image, Which Sits In Aws

Its a setting on the autoscaler:

2 months ago

0 [Pipeline] Am I Right In Saying A Pipeline Controller Can’T Include A Data-Dependent For-Loop? The Issue Is Not Spinning Up The Tasks, It’S Collecting The Results At The End. I Was Trying To Append The Outputs Of Each Iteration Of The For-Loop And Pass Th

Hi ReassuredOwl55 , can you please elaborate on your use case or exactly what you're trying to achieve?

2 years ago

0 Hi, Iam Trying To Get Clearml Dashboard With The Ec2 Instance But After Adding The Security Groups And Ssh Key To Instance Its Just Giving Below Error. How To Add Authentication To Ec2 Instance For Loading My Projects

Hi @<1536881167746207744:profile|EnormousGoose35> , can you take a look at the webserver and apiserver container logs to see what errors are there?

2 years ago

0 Hi, When Using A Pipeline Decorator How Do I Add A Local Module? I Keep On Getting This Error Message:

@<1523701295830011904:profile|CluelessFlamingo93> , I'm not sure what you mean. Whenever you run a pipeline code (pipeline from decorators) if it's from a repository that repo will be logged. Where are you importing "train" from? What if you import entire package & point to the specific module?

2 years ago

0 `I'M Attempting My First Project In Clearml And Have A Few Questions. I'M Migrating My Video Encoding Pipeline In Python3.12 To Clearml Using The Decorators.: @Pipelinedecorator.Pipeline( Name="Run Video Encoder Pipeline", Project="Video Encoder

And from the error you get, like I mentioned it looks there is no services queue. I would check the logs on the agent-services container to see if you get any errors as this is the agent in charge of listening to the 'services' queue

8 months ago

0 Hi, Anyone Facing Ssh Issues While Queueing Tasks Into Clearml-Agents Installed In The Server

What issues are you facing?

6 months ago

0 Hello, Everyone! Is There A Clearml .Pixi Package Manager Support ?

Hi @<1736556881964437504:profile|HelplessFly7> , I don't think there is such an integration. Currently Poetry, pip and Conda are supported. I think you could make a PR on this for the clearml-agent

one year ago

0 Hi Everyone ! I Am Trying To Remotely Run A Code From A Github Repo Which Necessitates Some Installs (As Shown In The Picture). What Is The Best Way To Run These After Cloning The Git Repo ? Some Observations I Made:

Hi SwankyCrab22 ,

Regarding Task.init() did you try passing docker_bash_setup_script and it didn't work? Because according to the docs it should be available with Task.init() as well. Also, after the Task.init() you can use the following method:
https://clear.ml/docs/latest/docs/references/sdk/task#set_base_docker
to also add a docker_setup_bash_script in it's args.

Regarding running the script after the repository is downloaded - I'm not sure. But certainly...

3 years ago

0 Hi Everyone!! Is Clearml Multi Gpu Support. I Have 2 Machines. Each Machine Have 2 A100 Gpu. Do We Have Any Option To Run Workload In These Gpu. Same Time Like Virtually. Do We Have The Support??

None

one year ago

0 Clearml Is Distinctly Missing In The Curated Mlops Landscape.

Hi @<1523701504827985920:profile|SubstantialElk6> , thanks for the heads up 🙂

2 years ago

0 Hi Everyone!

Hi @<1820993248525553664:profile|DisturbedReindeer69> , I think you're looking for the --output-uri parameter in clearml-data create - None

7 months ago

0 Hi, I Cannot Show The Results (E.G., 'Console', 'Scalars' ). When Pressing The Scalars Button, Clearml Shows A 'Wait Circle' And Then I Get The Message "Failed To Get Scalar Charts". A Solution That Worked 2 Days Ago, F12 --> Applications Tab --> Storage

RotundSquirrel78 , do you have an estimate how much RAM the machine running ClearML server? Is it dedicated to ClearML only or are there other processes running?

3 years ago

0 Hi, Does Self-Hosted Clearml Implements Central Model Repository, Something Like

By the way, is there a specific functionality you're looking for?

3 years ago

0 Hi Everyone! I Created A Pipeline From One Block, Passed The Initial Parameters. Please Tell Me, Is It Possible To Make A New Launch Of The Pipeline, But With Different Parameters, Just Like In The Draft Mode Of Usual Experiments. Globally, I Want To Init

Hi @<1524560082761682944:profile|MammothParrot39> , I think you need to run the pipeline at least once (at least the first step should start) for it to "catch" the configs. I suggest you run once with pipe.start_locally(run_pipeline_steps_locally=True)

2 years ago

0 I Am Trying To Run A Task On An Agent For The First Time But I Am Running Into Some Things I Do Not Understand, I Hope Someone Can Help Me Out With This. I Got An Agent Running On Google Colab, But When I Clone A Task And Enqueue It From The Web Ui, I Ge

Regarding the packages issue:
What python did you run on originally - Because it looks that 1.22.3 is only supported by python 3.8. You can circumvent this entire issue by running in docker mode with a docker that has 3.7 pre-installed

Regarding the data file loading issue - How do you specify the path? Is it relative?

3 years ago

0 Hi, For Some Reason, Running A Training Locally And Remotely (With A Set Seed For Np, Torch, And Random) Using Deterministic Mode.. Returns Different Results...

Hi @<1546303277010784256:profile|LivelyBadger26> , how did you set the random seed? I think you can also disable ClearML's random seed override and set one with Pytorch

2 years ago

0 Hi, What Is The Best Way To Add New Data To The Uploaded Dataset. Currently, Uploaded Dataset Stored In Clearml Have This Structure.

Hi, how did you save the dataset so far?

3 years ago

0 What Is The Best Way To Set S3 As A Files Server? We Have A Clearml Deployment Without A Files Server, But After/During A Training Run Clearml.Metrics Always Fails Due To A Connection Error While Trying To Call <Url>:8081 (We Don'T Have 8081 Because Of

Can you add the api section of your clearml.conf and also a log of a task?

one year ago

0 Can I Somehow Change A Project Settings Via The Python Sdk?

Hi UnevenDolphin73 , is there a specific setting you're looking for?

2 years ago

0 Hi There! I'M Getting An Error Whenever Trying To Queue Experiments Using Conda Package Manager That Require Python>=3.10. Locally (Non-Queued) The Training Runs Just Fine. In The Ui I See The The Following Console Output:

GreasyLeopard35 , please try with the latest clearml-agent

3 years ago

Show more results