AgitatedDove14

Sure:
Dataset.create(..., use_current_task=True)This will basically attach/make the main Task the Dataset itself (Dataset is a type of a Task, with logic built on top of it)
wdyt ?

4 years ago

0 <no title>

Hi @<1673501379764686848:profile|VirtuousSeaturtle4>

What I dont get is that the example does not refer to a bucket path. What bucket path should I specify ?

you mean to store data?

one year ago

0 , This Is A Great Tool For Visualizing All Your Experiments. I Wanted To Know That When I Am Logging Scalar Plots With Title As Train Loss And Test Loss They Are Getting Diplayed As Train Loss And Test Loss In The Scalar Tab. I Wanted That The Title Shoul

make sense ?

5 years ago

0 Hi! I Am Using The Modelcheckpoint Callback From Tensorflow To Save The Best Model. When The Experiment Finishes If I Go On The Server To Experiment > Artifacts > Output Model I Can See The Model And Subsequently By Clicking On It The Weights. How Can I

Yes, that sounds like the issue, is the file actually there ?

4 years ago

0 Another Question, I Have Written A Code That Includes A Task Scheduler That Calls A Function. That Function Watches A Folder And If There Are Sufficient Images, It Creates And Publishes The Dataset, After Which It Clears The Folder. Problem, For Some Rea

VexedCat68

a Dataset is published, that activates a Dataset trigger. So if every day I publish one dataset, I activate a Dataset Trigger that day once it's published.

From this description it sounds like you created a trigger cycle, am I missing something ?
Basically you can break the cycle by saying, trigger only on New Dataset with a specific Tag (or create the auto dataset in a different project/sub-project).
This will stop your automatic dataset creation from triggering the "orig...

3 years ago

0 Hey, Would It Possible To Add An Option To Make

Hmm, not a bad idea 🙂
Could you please open a Git Issue, so it will not get forgotten ?
(btw: I'm not sure how trivial it is to implement, nonetheless obviously possible 😉

5 years ago

0 Is There A Way To Limit The Number Of Jobs/Tasks That Can Run Concurrently On The

Hi ElegantCoyote26 , in theory no limit, but that depends on how you spined the services queue agent:
https://clear.ml/docs/latest/docs/clearml_agent/clearml_agent_daemon
See services mode :

To limit the number of simultaneous tasks run in services mode, pass the maximum number immediately after the

--services-mode

option (e.g.

--services-mode 5

)

3 years ago

0 Hello, How To Change The File Path Of Task’S Artifacts. I’M Migrating Clearml-Server From Servera To Serverb. I’D Also Like To Move The File Storage From Minioa To Miniob As Well. Currently, All Artifacts Have Been Cloned To Miniob. But When I Go To Clear

VictoriousPenguin97 I'm not sure there is an easy solution, basically you have to edit both MongoDB (artifacts) and Elastic (think debug samples) 😞

3 years ago

0 Hi Folks! I'M Using

Let me check...

4 years ago

0 Hello, I'M Trying To Setup Clearml-Agents. We Have Our Own Docker Image With All Dependencies Pre-Installed And A Script To Wrap Up All Repos, Is There A Way To Completely Override The Docker Command To Create The Task? It'S Mounting Volumes That Aren'T

Hi ShakyJellyfish91
Check mount default here:
https://github.com/allegroai/clearml-agent/blob/e416ab526ba9fe05daa977b34c9e46b50fb214a0/docs/clearml.conf#L186
Is this what you are after, or do you actually want to change the start up script?

3 years ago

0 Is There A Way To Control How Many Parallel Connections Are Used When Downloading From

okay let's PR this fix ?

3 years ago

0 Hi All, Looking For Some Help When Executing Pipelines With Custom Docker Images. I Have A Component Defined And I Expect Its Python Runtime Environment To Be Managed By A Custom Docker Image (

Funny enough I’m running into a new issue now.

Sorry my bad, I thought have known 😉 yes it probably should be packages=["clearml==1.1.6"]
BTW: do you have any imports inside the pipeline function itself ? if you do not, then no need to pass "packages" at all, it will just add clearml

3 years ago

0 Thanks For Releasing This Awesome Experiment Manager! I Was Logging A Single Training Session On Multiple Gpus (Using Detectron2), And Torch.Mp Is Called For Each Gpu. This Creates A Separate Task In Trains For Each Gpu, And Only One Of The Tasks Has The

Thanks VexedKangaroo32 , this is great news :)

5 years ago

0 Hi! I'M Currently Saving A Dataframe With Predictions Inside The Task. To Do So, I Save A Dataframe As Pickle File In

Yep, that would do it ...
You can disable it with:
Task.init(..., auto_connect_frameworks={'scikit': False})

4 years ago

0 Hi, It Seems Like We Have A Bug In Metrics Reporting While Comparing Between Several Experiments (Under Scalars). The Loss Report Includes Only One Experiment Results While All The Other Metrics Show All Of Them. The Data Is Exist At Each Experiment, But

What's the trains-server version ?

5 years ago

0 Hi, I'M Trying To Run Task.Init Inside A Jupyter Notebook For The First Time (Used It A Lot Before In Normal Python Scripts), And I Get A Warning-

😞 It's working as expected for me...
That said I tested on Linux & pip,
Any specific req to test with? from the log I see this is conda on windows, are you using the base conda env or a venv inside conda?

4 years ago

0 Hello, I Would Like To Use Spot Instances Together With The Aws Autoscaler To Train Models With Pytorch/Ignite And I Am Wondering How To Support Interruptions During The Training (In Case The Instance Is Terminated By Aws). Is There Anything Already Built

JitteryCoyote63 no I think this is all controlled from the python side.
Let me check something

4 years ago

0 Hello Clearml Ppl

Actually it would be interesting to combine the two, feast is fully open-source and supported by the linux foundation, so I cannot see the harm in that.
wdyt?

3 years ago

0 Hello Community! Is There An Option To Only Download A Part Of A Dataset With .Get_Local_Copy()? I Imagine Something Like This, But I Can'T Find The Right Way To Do It.

"warm" as you do not need to sync it with the dataset, every time you access the dataset, clearml will make sure it is there in the cache, when you switch to a new dataset the new dataset will be cached. make sense?

4 years ago

0 What Could Be The Reason For Fail Status Of A Task That Seems To Have Completed Correctly? No Information In The Log Whatsoever

0.17.2 ?

4 years ago

0 When Clearml Converts A

Okay the type is inferred from the default value of the function step itself, that means that both:
data_frame = step_one(pickle_url, extra=1337)and
data_frame = step_one(pickle_url, 1337)Will pass extra as int .
That said if the default value of the argument is missing, it will revert to str
In order to use the type hints as casting hint, we actually need to improve the task.connect to support the type casting (they are stored)

3 years ago

0 Hi, I Was Some How Able To Get A Project Running Yesturday, However Now I Am Unable To Get It Running, I Keep Getting An Failed Getting Token Error

restart the notebook kernel ?

4 years ago

0 Hi, I Have Quite A Generic Question. Basically, I Am Picking Your Brains For Any Solution. Our Current Pipeline Has (Clearml-Data, Clearml And Seldon). We Were Looking For Some Workflow Orchestrator To Stitch Them Up. One Scenario:

DeliciousBluewhale87 fyi, the new version of the pipeline (hopefully pushed towards the end of this week), will allow you to more easily write steps as functions (not only as Tasks, as the current implementation)
Also check the new Trigger and Scheduler both intended to trigger these pipelines:
https://github.com/allegroai/clearml/blob/fe3c481c37e70881c44d67c1cf9bbce00a84747e/clearml/automation/scheduler.py#L457
https://github.com/allegroai/clearml/blob/fe3c481c37e70881c44d67c1cf9bbce00a8...

4 years ago

0 When Using Something Like Pdf2Image Which Requires Poppler (Which Can Be Installed With Conda), How Can I Ensure That The Task Can Run On An Agent Correctly? As Of Now It Doesn’T Know About Poppler

So on the ec2 instance (with the agent running), just install prior to running the agent:
apt-get install poppler-utils

4 years ago

0 <image>

No sure what O'm seeing here

4 years ago

0 Why Would Every Submitted Task Be Aborted Directly?

Nothing except that Draft makes sense feels like the task is being prepped and Aborted feels like something went wrong

Yes guess that if we call execute remotely, without a queue, it makes sense for you to edit it...
Is that the case TrickySheep9 ?
If it is I think we should change it to draft when it is not queued. sounds good to you guys ?

4 years ago

Show more results