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
PricklyRaven28
Moderator
15 Questions, 99 Answers
  Active since 10 January 2023
  Last activity 3 days ago

Reputation

0

Badges 1

97 × Eureka!
0 Votes
1 Answers
573 Views
0 Votes 1 Answers 573 Views
Also it’s weird to me that when running pipelines in debugging_pipeline that it will even show in the UI
one year ago
0 Votes
0 Answers
26 Views
0 Votes 0 Answers 26 Views
5 days ago
0 Votes
6 Answers
570 Views
0 Votes 6 Answers 570 Views
Hey
Hey 🙂 Working with ClearML pipelines, is there a way to change default directory artifacts to archive with tar instead of zip ? Speaking of the default retu...
one year ago
0 Votes
8 Answers
625 Views
0 Votes 8 Answers 625 Views
There is a problem starting from clearml 1.7.0 with python-fire from clearml import Task import fire def check(first): print(first) if __name__ == '__main__'...
one year ago
0 Votes
12 Answers
575 Views
0 Votes 12 Answers 575 Views
Hey, Is there a way to temporarily turn off clearml logging? I’m using pipeline and when developing i don’t wand them to be added to the UI and spam it. in w...
one year ago
0 Votes
0 Answers
604 Views
0 Votes 0 Answers 604 Views
Hey 🙂 I’m also using the AWS autoscaler with pipelines, and i can’t seem to make the intermediate steps output their results to S3 instead of the fileserver...
one year ago
0 Votes
2 Answers
523 Views
0 Votes 2 Answers 523 Views
Hi
Hi 🙂 I'm trying to figure out if i have a way to report pipeline-step artifact paths in the main pipeline task. (So i don't need to dig into steps to find t...
one year ago
0 Votes
1 Answers
55 Views
0 Votes 1 Answers 55 Views
Hi
Hi 🙂 How can i disable requirements install when an agent is starting a task? we use docker, so everything should already be installed
18 days ago
0 Votes
7 Answers
606 Views
0 Votes 7 Answers 606 Views
i think there is some bug with clearml==1.7.1. I’m working with pipelines and after updating to 1.7.1 all the kwargs are passed as None to the steps, when do...
one year ago
0 Votes
0 Answers
633 Views
0 Votes 0 Answers 633 Views
(using a local clearml server)
one year ago
0 Votes
30 Answers
535 Views
0 Votes 30 Answers 535 Views
Hey everyone 🙂 I’m trying to use a ClearML on prem for experiment visualization only, having some issues with multi GPU. It seems that clearml is creating a...
2 years ago
0 Votes
9 Answers
13 Views
0 Votes 9 Answers 13 Views
https://clearml.slack.com/archives/CTK20V944/p1713357955958089 Any idea about this?
4 days ago
0 Votes
30 Answers
547 Views
0 Votes 30 Answers 547 Views
If possible, i would like all together prevent the fileserver and write everything to S3 (without needing every user to change their config)
one year ago
0 Votes
62 Answers
21K Views
0 Votes 62 Answers 21K Views
Hey, We are using clearml 1.9.0 with transformers 4.25.1… and we started getting errors that do not reproduce in earlier versions (only works in 1.7.2 all 1....
one year ago
0 Votes
2 Answers
458 Views
0 Votes 2 Answers 458 Views
Hey
Hey 🙂 I'm using pipelines with decorators and trying to set a custom docker image to the pipeline itself which doesn't seem to accept a "docker=" argument l...
10 months ago
0 How Can I Ensure Tasks In A Pipeline Have The Same Environment As The Pipeline Itself? It Seems A Bit Counter-Intuitive That The Pipeline (Executed Remotely) Captures The Local Environment, But The Tasks (Executed Remotely) Do Not Use That Same Environmen

We also wanted this, we preferred to create a docker image with all we need, and let the pipeline steps use that docker image

That way you don’t rely on clearml capturing the local env, and you can control what exists in the env

one year ago
0 Hey Everyone

i get for one of the tasks, but then it fails because it seems that the fastai2 tensorboardcallback isn’t fit for distributed training (which i’m opening an issue for them now)

2 years ago
0 Hey Everyone

this
from fastai.callbacks.tensorboard import LearnerTensorboardWriter
doesn’t exist anymore in fastai2

2 years ago
0 Hey Everyone

but that means that there is no way to work with clearml+fastai2+multi gpu

2 years ago
0 Hey Everyone

maybe the fastai bindings in clearml package are outdated

2 years ago
0 Hey Everyone

you can get updates on the issue i opened
https://github.com/fastai/fastai/issues/3543

but i think the probably better solution would be to create a custom ClearML callback for fastai with the best practices you think are needed…

Or try to fix the TensorBoardCallback, because for now we can’t use multi gpu because of it 😪

2 years ago
0 Hey Everyone

but anyway, this will still not work because fastai’s tensorboard doesn’t work in multi gpu 😞

2 years ago
0 Hey Everyone

reduced to a small snippet
` from fastai.vision.all import *
from fastai.distributed import *
from clearml import Task
from fastai.callback.tensorboard import TensorBoardCallback
from wwf.vision.timm import timm_learner

task = Task.init(project_name='LIOR_TEST', auto_connect_arg_parser={'rank': False})
path = untar_data(URLs.PETS)

size = 460
batch_size = 32

dblock = DataBlock(blocks=(ImageBlock, CategoryBlock),
get_items=get_image_files,
get_y=lambda x: ...

2 years ago
0 Hey Everyone

i’m following this guide
https://docs.fast.ai/distributed.html#Learner.distrib_ctx

so you run it like this
python -m fastai.launch <script>

2 years ago
0 Hey Everyone

That’s how fastai distributes between multiple gpus, i’ll try to move the init

2 years ago
0 Hey Everyone

unrelated, i’m getting a crash, but seems related to fastai and not clearml

2 years ago
0 Hey Everyone

still getting 4 tasks (1 does have data in results)

2 years ago
0 Hey Everyone

any idea? 🙏

2 years ago
0 Hey Everyone

AgitatedDove14 . so if i understand correctly, what i can possibly do is copy paste the https://github.com/fastai/fastai/blob/master/fastai/launch.py code and add the Task.init there?

2 years ago
0 Hey Everyone

when u say use Task.current_task() you for logging? which i’m guessing that the fastai binding should do right?

2 years ago
0 Hey Everyone

and use that instead of the -m fastai.launch part ofcourse

2 years ago
0 Hey Everyone

i’ll check if it’s already initialized

2 years ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):

` args.py #504:
for k, v in dictionary.items():
# if key is not present in the task's parameters, assume we didn't get this far when running
# in non-remote mode, and just add it to the task's parameters
if k not in parameters:
self._task.set_parameter((prefix or '') + k, v)
continue

task.py #1266:
def set_parameter(self, name, value, description=None, value_type=None):
# type: (str, str, Optional[str], O...

one year ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):

This is the next step not being able to find the output of the last step

ValueError: Could not retrieve a local copy of artifact return_object, failed downloading 
one year ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):
Traceback (most recent call last):
  File "/tmp/tmpxlf2zxb9.py", line 31, in <module>
    kwargs[k] = parent_task.get_parameters(cast=True)[return_section + '/' + artifact_name]
KeyError: 'return/return_object'
Setting pipeline controller Task as failed (due to failed steps) !
Traceback (most recent call last):
  File "/usr/src/lib/clearml_test.py", line 69, in <module>
    pipeline()
  File "/opt/conda/lib/python3.10/site-packages/clearml/automation/controller.py", line 3914, in intern...
one year ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):

` from clearml.automation import PipelineDecorator
from clearml import TaskTypes

@PipelineDecorator.component(task_type=TaskTypes.data_processing, cache=True)
def run_demo():
from transformers import AutoTokenizer, DataCollatorForTokenClassification, AutoModelForTokenClassification, TrainingArguments, Trainer
from datasets import load_dataset

dataset = load_dataset("conllpp")

model_checkpoint = 'bert-base-cased'
lr = 2e-5
num_train_epochs  = 5
weight_decay =...
one year ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):

I'm getting really weird behavior now, the task seems to report correctly with the patch... but the step doesn't say "uploading" when finished... there is a "return" artifact but it doesn't exist on S3 (our file server configuration)

one year ago
0 Hey, We Are Using Clearml 1.9.0 With Transformers 4.25.1… And We Started Getting Errors That Do Not Reproduce In Earlier Versions (Only Works In 1.7.2 All 1.8.X Don’T Work):

@<1523701435869433856:profile|SmugDolphin23>
Hey 🙂
Any update?

We are having more issues with transformers and clearml in their new version.
The step that has transformers 4.25.1 isn’t able to upload artifacts.
If we downgrade transformers==4.21.3 it works

one year ago
Show more results compactanswers