CostlyOstrich36

0 Questions, 4212 Answers

Active since 10 January 2023

Last activity 2 years ago

Reputation

Answers 4212

0 Hi All! I Can'T Use Scalar Tab In All Experiments Due To Elastic Search Error:

Please run the following commands and share the results. Chances are that somehow the default mappings that we apply on the index creation were not applied to your events scalar index.

First run the following command

curl -XGET "localhost:9200/_cat/indices/events-training_stats_scalar-*"

And then for each of the returned indices run the following:

curl -XGET "localhost:9200/<index_name>/_mappings"

2 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 Is There A Robust Way (Using The Sdk And Not The Ui) To Add Tags To Task Regardless Of Where It Is Executed?

I think you can get the task from outside and then add tags to that object

3 years ago

0 Hi Everyone! I'Ve Noticed That If I Run An Experiment And It Fails, The Clearml Agent Will Delete All Datasets That Have Been Downloaded During The Run. Is It Correct Behavior? How Can I Force The Agent To Preserve Such Datasets?

ExcitedSeaurchin87 , Hi 🙂

I think it's correct behavior - You wouldn't want leftover files flooding your computer.

Regarding preserving the datasets - I'm guessing that you're doing the pre-processing & training in the same task so if the training fails you don't want to re-download the data?

3 years ago

0 Hey Everyone, Do I Understand Correctly That There Is No Way To Enable "Access Rules" And "Users & Groups" Tabs In Webui For My Self-Hosted Clearml Server Using Some Sort Of Configuration File Since This Functionality Is Simply Not In The Free Open Source

Hi UpsetSheep55 ,

Permissions feature is indeed exists only in the enterprise version. There are no examples for this since this an enterprise only feature.

3 years ago

0 Is There Any Way (Or Are There Any Plans) To Include Some View For Datasets In The Webui? One That Is Detached From The Generating Task?

As far as I know Hyper-datasets also support csv/tabular data quite well 🙂

3 years ago

0 Hi, I Am Running A Pipeline From Decorators. The Pipeline Runs Fine. Then I Try To Clone It By Clicking The (Successful) Run And Launching. The Pipeline Fails Immediately With The Error

Hi RoughTiger69 , how are you running the pipeline? Locally or on agents? How is the controller running?

3 years ago

0 Hi, What Would Be The Best Practice For Using The Datasets When Testing Different Features? Should I Just Add New Dataset Each Time I Try New Feature To The Dataset And Specify The Previous One As Parent?

Hi @<1566596960691949568:profile|UpsetWalrus59> , yes that sounds like a good way 🙂

2 years ago

0 Hi, I'M Using Clearml Over Aws And I Can See That My Ec2 Instance Disk Space Is Filling Up Really Fast. I First Started With 100Gb Disk Space And Once It Filled Up I Increased The Size To 150Gb Which Is Soon To Be Filled Up As Well. I Wanted To Check If T

Can you connect directly to the instance? If so, please check how large /opt/clearml is on the machine and then see the folder distribution

3 years ago

0 I Assume I Can Ask A Question Here. The Clearml Orchestrator Looks Interesting. But The Website Suggests That K8S Is Required. We Have A Linux Training Box (Lambdabox) Where We Want To Run Training. Can We Place The Clearml Orchestrator Agent On The M

RobustFlamingo1 , I think this is because you looked at 'Orchestrate for DevOps' and not 'Automate for Data Scientist'. If you switch to the other option you will see no K8S is required 🙂

I am guessing that the use-case shown there would be more what you're looking for. The K8S is something for larger scale deployments when the DevOps guys set up the system to run on K8S cluster

3 years ago

0 Hi, What Would Be The Recommended Way To Add/Track Arbitrary Models To/With Outputmodels? Currently Hacking It By Using Joblib Dump And Subsequently Deleting Unwanted "Local" Files. Arbitrary In This Case Just Extensions To Some Scikitlearn Classes.

If you set Task.init(..., output_uri=<PATH_TO_ARTIFACT_STORAGE>) everything will be uploaded to your artifact storage automatically.
Regarding models. I to skip the joblib dump hack you can simply connect the models manually to the task with this method:
https://clear.ml/docs/latest/docs/references/sdk/model_outputmodel#connect

3 years ago

0 We Are Getting "Fetch Experiment Failed" In Ui In One Of Projects. Other Projects Are Ok. In

Are you running a self deployed server? What is the version if that is the case?

3 years ago

0 Hi! Im Trying Pipelines With Data Creating And Model Training.

Hi AbruptHedgehog21 , it looks like you need to parameters.dataset_id on step data_creation

2 years ago

0 Hi. We Have Task That Generate Trained Model In

Yep 🙂

2 years ago

0 Hi, I Would Like To Understand How I Can Set The Pip Cache Location For My Agent, I Thought That I Already Had The Right Setting With

Can you add a bit more from the log for more context as well?

3 years ago

0 Hey Everyone Is There An Option In The Ui/Scalars Wizard To Have Metrics Aggregated Into Single Graph, E.G The Ability To See

Hi @<1523701842515595264:profile|PleasantOwl46> , I suggest opening a GitHub feature request for this 🙂

3 months ago

0 Is There A Way To Configure A Clearml-Agent So That It Shutdown The Server After It Has Been Idle For A Certain Time Period? We Are Using Gpu Resources From A Provider That Autoscaling Doesn'T Support (Such As Sagemaker Training Jobs).

Any specific reason not to use the autoscaler? I would imagine it would be even more cost effective

2 years ago

0 Hello, Deploy Clearml In Docker And Train A Yolov8 Model, Finish Train, In Plots No Pics Show,So Why

Are you using the community server or are you using the open source and self hosting?

one year ago

0 Hi, Is There A Way To Get Clearml.Conf Using Clearml Sdk? Tia

What is the use case of accessing clearml.conf during runtime?

2 years ago

0 Hello! How Can I Save And Resume Studies With Optune? If Best Quality Have Not Reached Or Another Reasons. Optune Have Capabilities For Resuming Studies.

Hi!
I believe you can stop and resume studies by adding these actions to your script:

Add save points via joblib.dump()
and connect them to clearml via clearml.model.OutputModel.connect()

Then, when you want to start or resume a study, load latest study file via joblib.load() and connect to clearml with http://clearml.model.InputModel.co nnect()

This way you can stop your training sessions with the agent and resume them from nearly the same point

I think all the required references are h...

4 years ago

0 [Clearml-Session Question] Why Does Jupyter Lab Have Only Token In Url While Code-Server Doesn’T?

Hi @<1524922424720625664:profile|TartLeopard58> , can you elaborate on what do you mean by code-server?

2 years ago

0 I Am Experiencing Performance Issues With Using Clearml Together With Pytorch Lightning Cli For Experiment Tracking. Essentially What We'Re Doing Is Fetching The Logger Object Through Task.Get_Logger() And Then Using The Reporting Methods. However, It Ad

Hi SoreHorse95 ,

Does ClearML not automatically log all outputs?

Regarding logging maybe try the following setting in ~/clearml.conf sdk.network.metrics.file_upload_threads: 16

3 years ago

0 Hi, Thanks A Lot For Your Product ! I Have A Question About Possible Functionality. Is It Possible Now To Group Graphics In "Scalars" Tab In Dropdown Groups As It Done In Tensorboard. I'Ll Try To Explain What I Mean, Let Say That I Am Solving Object Detec

Hi SillyGoat67 ,

Hmmm. What if you run these in separate experiments and each experiment reports it's own result? This way you could use comparison between experiments to see the different results grouped together.

Also you can report different scalars for the same series so you can see something like this:

3 years ago

0 Hello, I'M Getting This Weird Error From Time To Time When Running A Pipeline, It Add My Tasks As Drafts But Never Launch Them, When I Checked The Logs, I See The Following ;

Can you gain access to the apiserver logs?

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 Hi I Am Encountering Some Difficulties While Trying To Run The Examples Of The Clearml Documenation (E.G.

Also can you provide the full log for better context?

3 years ago

0 When I Upload And Publish Data To Clearml-Data, It Says Successful. Now When I Try To Get It Using Id, I Get The Following Error. Error: __Get_Tasks() Got Multiple Values For Keyword Argument 'Tags'. This Occurs Both In Cli And In Python.