Hi SuperiorCockroach75 , yes you can use any docker image you want 🙂

@<1523701070390366208:profile|CostlyOstrich36> While running pipeline.py for the asteroid_example.
It worked for this: pipe.start_locally(run_pipeline_steps_locally=True)

But it's in pending state only while running with pipe.start().

ClearML Task: created new task id=0f27eae0f8a34605870ebbf41a2ad128
ClearML results page: 

ClearML pipeline page: 

Switching to remote execution, output log page 

Terminating local execution process

python:3.7
Status: Downloaded newer image for python:3.7
docker: Error response from daemon: could not select device driver "" with capabilities: [[gpu]].

@John C. where to configure if agent is not having gpu?

@<1533159639040921600:profile|JoyousReindeer30> , you need to run the agent with the -- cpu-only tag as well 🙂

Agent is also running:
clearml-agent daemon --queue default services --foreground --docker python:new --cpu-only

but agent setting stucked at this pip warning

Hi @<1533159639040921600:profile|JoyousReindeer30> , the pipeline controller is currently pending. I am guessing it is enqueued into the services queue. You would need to run an agent on the services queue for the pipeline to start executing 🙂

pipeline gets created but execution not happening...