UpsetBlackbird87

3 Questions, 8 Answers

Active since 10 January 2023

Last activity one year ago

Reputation

Badges 1

8 × Eureka!

Questions 3
Answers 8

0 Votes

2 Answers

825 Views

0 Votes 2 Answers 825 Views

Hi, I Have A Question About Queue Management Of Clearml Agents. I Am Still A Beginner To Clearml And Still Discovering The Potential It Has And As Of Now It Has Amazed Me With It Versatile Features

Hi, I have a question about queue management of ClearML Agents. I am still a beginner to ClearML and still discovering the potential it has and as of now it ...

mlops

3 years ago

0 Votes

13 Answers

1K Views

0 Votes 13 Answers 1K Views

Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

Hi I’m trying out pipeline controller from tasks. I was not able to understand why my code results in just one task(the first one) in the pipeline. pipeline ...

clearml

2 years ago

0 Votes

2 Answers

1K Views

0 Votes 2 Answers 1K Views

Hi I Was Running An Hyperparameter Optimization Task Using The Optuna Optimizer And Even Though The Hyperparameteroptimizer’S Argument Is Set To

Hi I was running an Hyperparameter Optimization task using the Optuna Optimizer and even though the HyperParameterOptimizer’s argument is set to max_number_o...

clearml

2 years ago

0 Hi, I Have A Question About Queue Management Of Clearml Agents. I Am Still A Beginner To Clearml And Still Discovering The Potential It Has And As Of Now It Has Amazed Me With It Versatile Features

Thx for your response Martin.

3 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

CostlyOstrich36 The commented section is the 3 layer part which I am not currently using due to the lag
` # def update_datasets(dataset_projects, parents, is_ppg=False):

current_steps = []

for dataset_project in dataset_projects:

dataset_project_split = dataset_project.split('-')

campaign_num = dataset_project_split[2]

parameters = {

'Args/XXX': args.XXX_template.replace('*', campaign_num),

'Args/XXX': args.XXX_template....

2 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

For some reason my code results in one node even though I think the logic above should result in a bipartite graph

2 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

Here is my log after task execution
**Deleted**

2 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

AgitatedDove14 CostlyOstrich36 [FYI] I’m not sure if this is an optimization issue in ClearML or my computer’s issue but when I make the pipeline a tri-partite graph (add another parent layer) the DAG becomes weird (I think because of the lag) and the web app lags so much until I delete this pipeline in the web app a (I can’t interact will any nodes in the graph and switching between tasks takes a long time)

2 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

CostlyOstrich36 This resulted in a bipartite graph that I expected but why? 😕

2 years ago

0 Hi I’M Trying Out Pipeline Controller From Tasks. I Was Not Able To Understand Why My Code Results In Just One Task(The First One) In The Pipeline.

AgitatedDove14 Thx for the clear explanation!

2 years ago

0 Hi I Was Running An Hyperparameter Optimization Task Using The Optuna Optimizer And Even Though The Hyperparameteroptimizer’S Argument Is Set To

Thx so much 👍

2 years ago