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
Answered
Hey, I Am New With Clearml And Need Some Help

Hey, I am new with clearML and need some help 🙂

I am trying to run very simple pipeline inside docker container. I followed the documentation, created new queue ("docker-cpu-queue") and agent. It seems to work fine, I am able to run task inside docker (checked docker ps, and docker run). Tasks are completed successfully, but when I run pipeline, got Running status, but nothing happens. Here is log from docker

Starting Task Execution:

ClearML results page: 

ClearML pipeline page: 

2024-05-17 08:54:32,622 - clearml.util - WARNING - 2 task found when searching for `{'project_name': 'pipeline_example', 'task_name': 'Pipeline step 1', 'include_archived': True, 'task_filter': {'status': ['created', 'queued', 'in_progress', 'published', 'stopped', 'completed', 'closed']}}`
2024-05-17 08:54:32,622 - clearml.util - WARNING - Selected task `Pipeline step 1` (id=e758eb0f7f464c30a9bc61d32995f703)
2024-05-17 08:54:32,721 - clearml.util - WARNING - 2 task found when searching for `{'project_name': 'pipeline_example', 'task_name': 'Pipeline step 2', 'include_archived': True, 'task_filter': {'status': ['created', 'queued', 'in_progress', 'published', 'stopped', 'completed', 'closed']}}`
2024-05-17 08:54:32,721 - clearml.util - WARNING - Selected task `Pipeline step 2` (id=b433826ebe3a48eb96a235bd7b7ba6b9)
Launching the next 1 steps
Launching step [task_one]
Launching step: task_one
Parameters:
None
Configurations:
{}
Overrides:
{}

Here are my steps:

step1:

from clearml import Task

task = Task.init(project_name="pipeline_example", task_name="Pipeline step 1")
task.execute_remotely(queue_name="docker-cpu-queue")

print('Hello from task one!')

step2:

from clearml import Task

task = Task.init(project_name="pipeline_example", task_name="Pipeline step 2")
task.execute_remotely(queue_name="docker-cpu-queue")

print('Hello from task 2')

pipeline_controller:

from clearml import Task
from clearml.automation import PipelineController

def pre_execute_callback_example(a_pipeline, a_node, current_param_override):
    # type (PipelineController, PipelineController.Node, dict) -> bool
    print(
        "Cloning Task id={} with parameters: {}".format(
            a_node.base_task_id, current_param_override
        )
    )
    # if we want to skip this node (and subtree of this node) we return False
    # return True to continue DAG execution
    return True


def post_execute_callback_example(a_pipeline, a_node):
    # type (PipelineController, PipelineController.Node) -> None
    print("Completed Task id={}".format(a_node.executed))
    # if we need the actual executed Task: Task.get_task(task_id=a_node.executed)
    return


# Connecting ClearML with the current pipeline,
# from here on everything is logged automatically
pipe = PipelineController(
    name="Pipeline demo", project="pipeline_example", version="0.1", add_pipeline_tags=False
)

pipe.set_default_execution_queue("docker-cpu-queue")

pipe.add_step(
    name="task_one",
    base_task_project="pipeline_example",
    base_task_name="Pipeline step 1",
)

pipe.add_step(
    name="task_two",
    parents=["task_one"],
    base_task_project="pipeline_example",
    base_task_name="Pipeline step 2",

    pre_execute_callback=pre_execute_callback_example,
    post_execute_callback=post_execute_callback_example,
)

# pipe.start_locally()
#pipe.start_locally(run_pipeline_steps_locally=False)
pipe.start(queue="docker-cpu-queue")

print("done")

Any idea?

  
  
Posted one month ago
Votes Newest

Answers 9


This is the env variable you're looking for - CLEARML_AGENT_SKIP_PYTHON_ENV_INSTALL

None

  
  
Posted one month ago

Hi @<1702492411105644544:profile|YummyGrasshopper29> , it looks like the controller is running, but is there any agent listening to where the tasks are being pushed?

  
  
Posted one month ago

Yes, there is one agent. As I said, I am able to execute task, but have problem with pipeline

  
  
Posted one month ago

Thanks for very fast replay!

  
  
Posted one month ago

Ok, thanks for explanation. So pipeline controller is in Running state, while task 1 is in pending state. The solution will be to add one more agent?

  
  
Posted one month ago

tasks being the steps of the pipeline itself

  
  
Posted one month ago

Think of it this way. You have the pipeline controller which is the 'special' task that manages the logic. Then you have the pipeline steps. Both the controller and the steps need some agent to execute them. So you need an agent to execute the controller and also you need another agent to run the steps themselves.

I would suggest by clicking on 'task_one' and going into full details. My guess it is in 'enqueued' state probably to the 'default' queue.

  
  
Posted one month ago

It works! Thanks man, you save my day!!

  
  
Posted one month ago

One more question 🙂
How can I force clearML not to install requirements before running task? (already have everything installed on docker machine)

  
  
Posted one month ago
114 Views
9 Answers
one month ago
one month ago
Tags
Similar posts