CostlyOstrich36

0 Questions, 4175 Answers

Active since 10 January 2023

Last activity 2 years ago

Reputation

Answers 4175

0 Hi All, I Have A Problem That I Couldn'T Figure On My Own ,I Will Be Happy For Help On This Issue! I Run On Linux With A "Father" Script That Creates A Subprocess, Calls Task.Init To Create A Task, Then Save Some Metrics And Then Close The Task (Few Time

I'll try and see if it reproduces on my side, thanks! 🙂

3 years ago

0 But For Example The Pytorch Versions Would Be Different For The Two, Macbook Is Running A Different Version And My Gpu(An Nvidia Machine) An Another Version, How Can I Make Sure They Can Both Run Seamlessly Without Having Any Compatibility Issue?

I'd suggest using the agent in --docker mode

9 months ago

0 Hello, I'M Trying To Get The Pipeline Progress, But I Can'T Get It Properly. I'Ve Set The Progress Via

Hi FancyWhale93 , do you have a snippet that reproduces this?

2 years ago

0 I Wanted To Ask, I'M Versioning My Data Using Clearml Data. And I'Ll Have A Training Task With Clearml Task. My Question Is, Does Clearml Keep Track Of The Data Versions Fetched From Clearml Data? Basically I Want To See How Much Of Tracking And Informati

VexedCat68 , that's a good question! I'm not sure that ClearML keeps track of that, I need to check on that.

However, I think a neat solution could be using the datasets as task configuration parameters. This way you can track which datasets were used and you can set up new runs with different datasets.

3 years ago

0 Hello Everybody, I Am Using Clearml Self-Hosted. Is There A Way To Run Multiple Clearml Instances (Multiple Webservers, Apiservers And Fileservers) But With A Shared Databases (Elastic, Mongo)? Elasticsearch Requires A Lot Of Ram And It Would Be An Overki

Role based access controls are supported only in Scale/Enterprise versions.

2 years ago

0 Hey Everyone! First Post Here. My Issue Is The Python_Binary I Set In The .Conf File Doesn'T Seem To Get Used By The Task I'M Running. The Task'S Output Confirms

Hi @<1618780810947596288:profile|ExuberantLion50> , what happens if you set the python binary path itself? Also, any specific reason you're not running in docker mode?

one year ago

0 Hi! Which Method Is Used To Delete Outputmodel From The Storage?

How are you saving your models? torch.save ("<MODEL_NAME>") ?

3 years ago

0 Hi All, I Was Using The Cleaning Script, But It'S (I Think The Elasticsearch, Since It'S A Java Process) Consuming A Lot Of Memory. So Much That It'S Blocking The Server'S Access. The Cleaning Script Is Running On The Services Queue On The Server. Can I R

Hi @<1577468611524562944:profile|MagnificentBear85> , are you running the services agent in docker mode or venv?

7 months ago

0 Hi All - I Am Using The Experiment Tracker And I See That Scalars View Is Not Updated After Some Time? In The Console View I See That My Training Job Terminated At Step 708000, Having Logged Scalar Metrics Every 4 Steps, However In The Scalars Pane I See

Hi @<1655744373268156416:profile|StickyShrimp60> , do you have any code that can reproduce this behavior?

one year ago

0 Hi, What Would Be The Correct Way To Move An Existing Self-Hosted Clearml Environment (In Our Case Only Experiments. With Artifacts, Models, Debug Samples Etc.) From One Server To Another, With A Different Url And Possibly Different Port Numbers? Is Such

Hi @<1558986867771183104:profile|ShakyKangaroo32> , you can do it but keep in mind that models/artifacts/debug samples are all referenced as links inside mongo/ES, you'd have to migrate the databases for that

9 months ago

0 Hi, I’M Trying To Bypass Task’S “Installed Packages” And Require An Installation Of My Repo’S

StickyCoyote36 , Hi!

Is the agent ignoring the requirements.txt and only installing the "installed packages" in the task or installing them both?

3 years ago

0 Hi Everyone, I Have A Problem With My

What are the requirements specific in the experiment UI?

3 years ago

0 Are Nested Pipeline Component Supported? E.G. Having A Component Call Other Pipeline Components? I Tried To Run This Snippet:

Hi @<1570220858075516928:profile|SlipperySheep79> , nested pipelines aren't supported currently. What is the use case that you need it for?

2 years ago

0 Hi, I'Ve Recently Started Using The

Hi @<1817731748759343104:profile|IrritableHippopotamus34> , are you only using StorageManager or also the Task object? Do you have a code snippet that reproduces this behaviour?

5 months ago

0 Hi All, I Use Autoscalers In My Training Configuration, And I Have An Issue With Them.

Hi @<1708653001188577280:profile|QuaintOwl32> , you can set some default image to use. My default for most jobs is nvcr.io/nvidia/pytorch:23.03-py3

one year ago

0 Hi Everyone, I Am New To Clearml And Am Trying To Use Clearml Agent With A Docker Image For Run A Pipeline Remotely. My Agent Is Running And It Is Able To Pick Up The Tasks From The Queue. It Starts Running The Task Inside Docker. However, It Exits The T

What actions did you take exactly to get to this state?

one year ago

0 Quick Question: Is There A Way For A Task That Is Executing Remotely To Find Out Which Clearml Queue It Is In Or Was In?

Hi @<1523702496097210368:profile|ScantChimpanzee51> , that information is saved on the task object. You can fetch it either with the API or the SDK

2 years ago

0 Hi, I Am Using Clearml On A New Server, And Task.Init() Crashes When I Run It From (Remote) Pycharm But Ok When I Run From A Commandline. Running On Remote Pycharm Is Ok For Me On Another Server. The Error That I Get Is:

you can find the different cache folders that clearml uses in ~/clearml.conf

3 years ago

0 Hi Folks, Is It Possible To Disable Automatic Package Detection And Just Use Requirements.Txt? I Have Different Local And Remote (Agent) Environments So I'D Like To Keep Those Two Separate. I'Ve Tried Using Task.Set_Packages("Requirements_Remote.Txt") Aft

You can do that too, yes

one year ago

VexedCat68 , correct. But not only arg parse. The entire configuration section 🙂

3 years ago

0 Hey All, I'M Running A Task Inside Docker, And One Of The Functions Call: Latest_Dataset = Dataset.Get(Dataset_Project=Dataset_Project]) A I Get: Raise Valueerror('Could Not Load Dataset Id={} State'.Format(Task.Id)) Valueerror: Could Not Load Dataset

I think you need to do latest_dataset = Dataset.get(dataset_id=<DATASET_ID>)

3 years ago

0 Hi, I'M Looking For Documentation On Gcp Autoscalers. When I Search On The Docs Site, It Shows Me The Aws Autoscaler But Not The Gcp One. Can Someone Point Me To The Right Docs Page? Thanks!

Hi JumpyPig73 ,

It appears that only the AWS autoscalar is in the open version and other autoscalars are only in advanced tiers (Pro and onwards):
https://clear.ml/pricing/

3 years ago

0 Hi! We'Re Trying To Orchestrate An Experiment Through The Clearml Ui With A Remote Agent On K8S. Our Training Job Requires A Package Only Available On Our Internal Pypi Server. How Should We Go About This Because The Job Keeps Failing With:

Hi CostlyFox64 ,

Can you try configuring your ~/clearml.conf with the following?
agent.package_manager.extra_index_url= [ "https://<USER>:<PASSWORD>@packages.<HOSTNAME>/<REPO_PATH>" ]

3 years ago

0 Hello Everyone, I Have A Small Question For The

Hi @<1582179652133195776:profile|LudicrousPanda17> , I suggest doing a similar filtering in the UI with dev tools open (F12) and see what is sent by the web UI 🙂

2 years ago

0 While Running A Pipeline By Decorator, I Am Getting This Error

I would suggest googling that error

2 years ago

0 Hi Everyone, I’M Looking To Implement Authentication For My Self-Hosted Clearml Server. Specifically, I Noticed The Option To Enable Web Login Authentication, Which I Assume Only Secures Access To The Clearml Web Interface. However, I’D Like To Know:

Hi @<1695969549783928832:profile|ObedientTurkey46> , this is supported in the Scale/Enterprise licensees of ClearML (external IdP support). API access is always done using credentials.

9 months ago

0 Hello, Would It Be Possible To Add To The Output Table Of

Hi @<1648134232087728128:profile|AlertFrog99> , not sure I understand. Can you please elaborate on your use case?

one year ago

0 Hi, Can I Provide A Direct Url For Pytorch Wheel Through Clearml.Conf File Or Somewhere Else? I'M Having A Following Issue.

@<1664079296102141952:profile|DangerousStarfish38> , are you running different python versions on the different machines? Remote vs local

one year ago

0 Hi, I Have A Question. I Am Trying To Create A Pipeline Using The

Hi @<1736919317200506880:profile|NastyStarfish19> , the services queue is for running the pipeline controller itself. I guess you are self hosting the OS?

one year ago

0 Hi, I Have A Question. I Am Trying To Create A Pipeline Using The

Then the services agent should be part of it. There also should be a 'services' queue that by default should listen to the services queue

one year ago

Show more results