Hi guys, sorry, this thread was some reason forgotten,

pipeline I ran but during its execution the overrides (which is filled) is not taken into account and I have my default params.

Could this be the same issue as this one?
None

Yeah that sounds like it, let me check if we can quickly reproduce

FunnyAlligator17 can you provide the full log ?

yes let me know if it is enough because I needed to remove a lot of internal stuff (logs are associated to snippet above):

Sorry I need the full log ... feel free to DM it to me

I am using:

ClearML SDK Version: clearml==1.12.0
ClearML Server Version (Only for self hosted): 1.9.2-317

Not sure but kinf of. I also want to clone the pipeline and change the hyperparameters but not from one specific task more like globally on my pipeline but same conclusion as him ( it is running on default parameters, given when pipeline was 1st run ), I am using from_functions .
Not clear for me what you mean by running the cloned pipeline with an agent (maybe the answer is inside my snippet above)?

Hi FunnyAlligator17 , what SDK version are you using?

👀 following.
I have much the same issue, and it's mission-critical that I resolve it soon.

Hello SuccessfulKoala55 , just try it and still no success, same error as before Error executing job with overrides: [] even if inside Configuration/Hyperparameters/Args overrides is correctly fill

Hi FunnyAlligator17 , can you please try with the new ClearML SDK v1.12.2rc0?