Examples: query, "exact match", wildcard*, wild?ard, wild*rd
Fuzzy search: cake~ (finds cakes, bake)
Term boost: "red velvet"^4, chocolate^2
Field grouping: tags:(+work -"fun-stuff")
Escaping: Escape characters +-&|!(){}[]^"~*?:\ with \, e.g. \+
Range search: properties.timestamp:[1587729413488 TO *] (inclusive), properties.title:{A TO Z}(excluding A and Z)
Combinations: chocolate AND vanilla, chocolate OR vanilla, (chocolate OR vanilla) NOT "vanilla pudding"
Field search: properties.title:"The Title" AND text
Profile picture
PanickyMoth78
Moderator
33 Questions, 165 Answers
  Active since 10 January 2023
  Last activity 13 days ago

Reputation

0

Badges 1

164 × Eureka!
0 Votes
7 Answers
566 Views
0 Votes 7 Answers 566 Views
Hi. I am experimenting with clearml.Dataset and encountering an error. LockException: [Errno 11] Resource temporarily unavailable In my experiment, I make a ...
one year ago
0 Votes
30 Answers
590 Views
0 Votes 30 Answers 590 Views
Hi. I'd like to try the GCP autoscaler. What permissions does the service account that I provide to clearml need? (and what GCP API should I enable in the GC...
one year ago
0 Votes
7 Answers
567 Views
0 Votes 7 Answers 567 Views
Hi. I have a problem accessing repo code in pipeline components running in an AWS autoscaler (first attempts at doing this) My local clearml.conf file has ag...
one year ago
0 Votes
2 Answers
570 Views
0 Votes 2 Answers 570 Views
Hi. I've noticed that my clearml.conf has both: agent.git_user="" agent.git_pass=""and agent { ... git_user: "" git_pass: "" ... }What's the difference? Shou...
one year ago
0 Votes
22 Answers
736 Views
0 Votes 22 Answers 736 Views
I started two pipelines (using AWS autoscaler in app.clear.ml ). The pipelines ran concurrently, using the same pipeline code. Both failed in the same compon...
one year ago
0 Votes
3 Answers
589 Views
0 Votes 3 Answers 589 Views
Hi. Shoulf this command succeed in the presence of project lavi-testing and absence of dataset tmp_datset within it? from clearml import Dataset tmp_dataset ...
one year ago
0 Votes
2 Answers
637 Views
0 Votes 2 Answers 637 Views
I am using the AWS autoscaler and I wish to set my files server to be gs. I tried to do so by having this in the ADDITIONAL CLEARML CONFIGURATION window: api...
one year ago
0 Votes
25 Answers
682 Views
0 Votes 25 Answers 682 Views
Autoscaler parallelization issue: I have an AWS Autoscaler set up with a resource that has a max of 3 instances assigned to the default queue I've given it a...
one year ago
0 Votes
11 Answers
589 Views
0 Votes 11 Answers 589 Views
Hi. I have a few questions about the snippet attached re-running this code produces the same printouts... I chose 47 out of 100 in the pipeline ... I chose 8...
one year ago
0 Votes
13 Answers
577 Views
0 Votes 13 Answers 577 Views
Another question on the topic of how a remote execution of a pipeline kills the calling process (previously discussed https://clearml.slack.com/archives/CTK2...
one year ago
0 Votes
1 Answers
556 Views
0 Votes 1 Answers 556 Views
suppose I use a pipeline decorator to define a pipeline: @PipelineDecorator.pipeline(name='my-pipeline', project='my-project', version='0.2') def my_pipeline...
one year ago
0 Votes
3 Answers
619 Views
0 Votes 3 Answers 619 Views
one year ago
0 Votes
7 Answers
602 Views
0 Votes 7 Answers 602 Views
Hi I'm looking into how clearml supports datasets and dataset versioning and I'm a bit confused. Is dataset versioning not supported at all in the non-enterp...
one year ago
0 Votes
16 Answers
738 Views
0 Votes 16 Answers 738 Views
Hi. Question about Dataset upload errors: When uploading a clearml.Dataset created with output_uri=" gs://lavi_test/datasets after adding 20 files of size 50...
gcp
one year ago
0 Votes
9 Answers
615 Views
0 Votes 9 Answers 615 Views
Hi. I have a question about pipelines and their generated dependency graphs. I took the code of the clearml pipeline from decorator example: https://github.c...
one year ago
0 Votes
2 Answers
633 Views
0 Votes 2 Answers 633 Views
Hi. Suppose I want to report on what my task has done by having it generate a markdown (.md) file with links to some "local" figure files. looking at the rep...
one year ago
0 Votes
14 Answers
669 Views
0 Votes 14 Answers 669 Views
Hi. I have a job that processes images and creates ~5 GB of processed image files (lots of small ones). At the end - it creates a clearml.Dataset and perform...
one year ago
0 Votes
8 Answers
609 Views
0 Votes 8 Answers 609 Views
Hi (again... sorry for asking so many questions) Question about using google cloud storage in a clearml agent running in AWS ec2 instance. my clearml.conf ha...
one year ago
0 Votes
3 Answers
583 Views
0 Votes 3 Answers 583 Views
Hi. First time user here 👋 I have experienced a problem following the getting started documentation. I opened an account on https://app.clear.ml/ I then fol...
one year ago
0 Votes
14 Answers
570 Views
0 Votes 14 Answers 570 Views
Bug? dataset name is ignored if use_current_task=True
one year ago
0 Votes
22 Answers
609 Views
0 Votes 22 Answers 609 Views
Hi. I'm encountering a problem with model.name At least, for models that where auto-magically uploaded. I see it in my own code but you can see it if you run...
one year ago
0 Votes
7 Answers
564 Views
0 Votes 7 Answers 564 Views
I have 5 unarchived pipeline runs that were defined with this decorator: @PipelineDecorator.pipeline( name="fastai_image_classification_pipeline", project="l...
one year ago
0 Votes
2 Answers
689 Views
0 Votes 2 Answers 689 Views
I have a training task that auto-magically saves a model for me to GCS task = Task.init( project_name=project_name, task_name=f"Image classification training...
one year ago
0 Votes
27 Answers
605 Views
0 Votes 27 Answers 605 Views
Hi. I'm running this little pipeline: from clearml.automation.controller import PipelineDecorator from clearml import TaskTypes @PipelineDecorator.component(...
one year ago
0 Votes
8 Answers
652 Views
0 Votes 8 Answers 652 Views
one year ago
0 Votes
20 Answers
635 Views
0 Votes 20 Answers 635 Views
task struck at task.flush(wait_for_uploads=True) : I've been running a model training task - a variation on this clearml dataset example: https://github.com/...
one year ago
0 Votes
8 Answers
701 Views
0 Votes 8 Answers 701 Views
one year ago
0 Votes
2 Answers
571 Views
0 Votes 2 Answers 571 Views
Hi. I'm using @PipelineDecorator.component to define a task from a function (to run in a pipeline) I'd like to get the task object within this function so th...
one year ago
0 Votes
1 Answers
625 Views
0 Votes 1 Answers 625 Views
one year ago
0 Votes
6 Answers
580 Views
0 Votes 6 Answers 580 Views
Is there some built-in way in clearml to trigger further action on task fail (or pipeline fail)?
one year ago
Show more results questions
one year ago
0 I Started Two Pipelines (Using Aws Autoscaler In App.Clear.Ml ). The Pipelines Ran Concurrently, Using The Same Pipeline Code. Both Failed In The Same Component Half-Way Though The Pipeline Run With:

Hey Alon,
See
https://clearml.slack.com/archives/CTK20V944/p1658892624753219
I was able to isolate this as a bug in clearml 1.6.3rc1 that can be reproduced outside of a task / app simply be doing get_local_copy() on a dataset with parents.

one year ago
0 I Started Two Pipelines (Using Aws Autoscaler In App.Clear.Ml ). The Pipelines Ran Concurrently, Using The Same Pipeline Code. Both Failed In The Same Component Half-Way Though The Pipeline Run With:

Another issue, may, or may not be related.
Running another pipeline (to see if I can reproduce the issue with simple code), it looks like the autoscaler has spun down all the instances for the default queue while a component was still running.
Both the pipline view and the "All experiment" view shows the component as running.
The component's console show that last command was a docker run command

one year ago
0 Hi. I Have A Question About Pipelines And Their Generated Dependency Graphs. I Took The Code Of The Clearml Pipeline From Decorator Example:

I think this should be a valid use of pipelines. for example - at some step I choose to sweep across several values of some parameter and the rest of the steps are duplicated for each value of that parameter.
The additional edges in the graph suggest that these steps somehow contain dependencies that I do not wish them to have.

one year ago
0 Hi. First Time User Here

In case anyone else is interested. We found two alternative solutions:
Repeating the first steps but from within a Docker container ( docker run -it --rm python:3.9 bash ) worked for me.alternatively
The example tasks (or at least those I've tried) that appear in the clear ml examples within a new workspace have clearml==0.17.5 (an old clearml version) listed in "INSTALLED PACKAGES". Updating the clearml package within the task to 1.5.0 let me run the clear-ml agent daemon lo...

one year ago
0 Hi There I'M Trying Out Clearml. I Saw Mention That Clearml Can Capture Tensorboard Output So I Tried It With This Little Script (Image Below). The Events File Is Filled, The Clearml Task Is Created, And Marked Complete However There Is Nothing In The Sc

here is the code in text if you feel like giving it a try:
import tensorboard_logger as tb_logger from clearml import Task task = Task.init(project_name="great project", task_name="test_tb_logging") task_tb_logger = tb_logger.Logger(logdir='./tb/run1', flush_secs=2) for i in range(10): task_tb_logger.log_value("some_metric", 42, i) task.close()

one year ago
0 Hi. I Am Experimenting With

I'm on clearml==1.6.3rc1

one year ago
0 Hi. Shoulf This Command Succeed In The Presence Of Project

That would be a better message however, I must have misunderstood the meaning of auto_create=True
I thought that flag made the get function into a "get-or-create"

one year ago
0 Another Question On The Topic Of How A Remote Execution Of A Pipeline Kills The Calling Process (Previously Discussed

on the same topic. What if (I were able to iterate and) I wanted the pipelines calls to be blocking so that the next pipeline executes only after the previous one completes?

one year ago
0 Another Question On The Topic Of How A Remote Execution Of A Pipeline Kills The Calling Process (Previously Discussed

You can have

parents

as one of the

@PipelineDecorator.component

args. The step will be executed only after all the

parents

are executed and completed

Is there an example of using parents some place? Im not sure what to pass and also, how to pass a component from one pipeline that was just kicked off to execute remotely (which I'd like to block on) to a component of the next pipeline's run

one year ago
0 Another Question On The Topic Of How A Remote Execution Of A Pipeline Kills The Calling Process (Previously Discussed

yes
here is the true "my_pipeline" declaration:
` @PipelineDecorator.pipeline(
name="fastai_image_classification_pipeline",
project="lavi-testing",
target_project="lavi-testing",
version="0.2",
multi_instance_support="",
add_pipeline_tags=True,
abort_on_failure=True,
)
def fastai_image_classification_pipeline(
run_tags: List[str],
i_dataset: int,
backbone_names: List[str],
image_resizes: List[int],
batch_sizes: List[int],
num_train_epochs: i...

one year ago
0 Another Question On The Topic Of How A Remote Execution Of A Pipeline Kills The Calling Process (Previously Discussed

nice, so a pipeline of pipelines is sort of possible. I guess that whole script can be run as a (remote) task?

one year ago
0 Another Question On The Topic Of How A Remote Execution Of A Pipeline Kills The Calling Process (Previously Discussed

I've also not figured out how to modify the examples above to wait for one pipline to end before the next begins

one year ago
0 Autoscaler Parallelization Issue: I Have An Aws Autoscaler Set Up With A Resource That Has A Max Of 3 Instances Assigned To The

erm,
this parallelization has led to the pipeline task issuing a bunch of:
model_path/run_2022_07_20T22_11_15.209_0.zip , err: [Errno 28] No space left on deviceand quitting on me.
my train_image_classifier_component is programmed to save model files to a local path which is returned (and, thanks to clearml, the path's contents are zipped uploded to the files service).

I take it that these files are also brought into pipeline tasks's local disk?
Why is that? If that is indeed what...

one year ago
0 Autoscaler Parallelization Issue: I Have An Aws Autoscaler Set Up With A Resource That Has A Max Of 3 Instances Assigned To The

Where was it running?

this message appears in the pipeline task's log. It is preceded by lines that reflect the storage manager downloading a corresponding zip file

I take it that these files are also brought into pipeline tasks's local disk?

Unless you changed the object, then no, they should not be downloaded (the "link" is passed)

The object is run_model_path
I don't seem to be changing it. I just pass it along from the training component to the evaluation compo...

one year ago
0 Autoscaler Parallelization Issue: I Have An Aws Autoscaler Set Up With A Resource That Has A Max Of 3 Instances Assigned To The

Note that the same models files were previously also generated by a non-paralelized version of the same pipeline without the out-of-space error but a storage manager was downloading zip files in that version as well (maybe these files were downloaded and removed as the object reference counts went to 0?)

one year ago
0 Autoscaler Parallelization Issue: I Have An Aws Autoscaler Set Up With A Resource That Has A Max Of 3 Instances Assigned To The

Two values:
`
@PipelineDecorator.component(
return_values=["run_model_path", "run_tb_path"],
cache=False,
task_type=TaskTypes.training,
packages=[
"clearml",
"tensorboard_logger",
"timm",
"fastai",
"torch==1.11.0",
"torchvision==0.12.0",
"protobuf==3.19.*",
"tensorboard",
"google-cloud-storage>=1.13.2",
],
repo="git@github.com:shpigi/clearml_evaluation.git",
repo_branch="main",
)
def train_ima...

one year ago
0 Autoscaler Parallelization Issue: I Have An Aws Autoscaler Set Up With A Resource That Has A Max Of 3 Instances Assigned To The

This example seems to suffice
Perhaps I should mention that I use gs as my files service ( files_server: gs://clearml-evaluation/ )
` from clearml.automation.controller import PipelineDecorator
from clearml import TaskTypes

@PipelineDecorator.component(
return_values=["large_file_path"], cache=False, task_type=TaskTypes.data_processing
)
def step_write(i: int):
import os

large_file_path = f"/tmp/out_path_{i}"
os.makedirs(large_file_path)
with open(f"{large_file_pa...
one year ago
Show more results compactanswers