[PWGHF] taskCharmPolarisation: Enhance the resolution of the ML scores outputs

[Mingze129]

No description provided.

[github-actions]

O2 linter results: ❌ 0 errors, ⚠️ 0 warnings, 🔕 5 disabled

[mfaggin]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

[vkucera]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

Absolutely agree.
Btw, what is the commented out configThnAxisMlPrompt doing there?

[Mingze129]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

As we checked this week for PbPb analysis, there are still too much background at low pt range (<5 GeV/c), and almost no signal peak at larger non-promtp cut. So we want a tighter cut on background.
I think it should be better to set configThnAxisMlBkg and configThnAxisMlNonPrompt as Configurables but now they are not. Maybe I can improved this.

[Mingze129]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

Absolutely agree. Btw, what is the commented out configThnAxisMlPrompt doing there?

The configThnAxisMlPrompt used to store the prompt scores, we move this now to reduce the output size.
But base on Xinye's suggestion, Sometimes the analysis works better on prompt cut instead of nonprompt cut. So we want to keep this for possible update in the future.

[mfaggin]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

As we checked this week for PbPb analysis, there are still too much background at low pt range (<5 GeV/c), and almost no signal peak at larger non-promtp cut. So we want a tighter cut on background. I think it should be better to set configThnAxisMlBkg and configThnAxisMlNonPrompt as Configurables but now they are not. Maybe I can improved this.

"I think it should be better to set configThnAxisMlBkg and configThnAxisMlNonPrompt as Configurables but now they are not. Maybe I can improved this."
What does it mean? These two are ConfigurableAxes which can be set as you want at runtime:

Are you saying that changing the values in Hyperloop does not give you the desired binning in the output?

[Mingze129]

a question from my side: is it really necessary? configThnAxisMlBkg and configThnAxisMlNonPrompt are Configurables, you should be able to set the values you want at runtime

As we checked this week for PbPb analysis, there are still too much background at low pt range (<5 GeV/c), and almost no signal peak at larger non-promtp cut. So we want a tighter cut on background. I think it should be better to set configThnAxisMlBkg and configThnAxisMlNonPrompt as Configurables but now they are not. Maybe I can improved this.

"I think it should be better to set configThnAxisMlBkg and configThnAxisMlNonPrompt as Configurables but now they are not. Maybe I can improved this." What does it mean? These two are ConfigurableAxes which can be set as you want at runtime:

Are you saying that changing the values in Hyperloop does not give you the desired binning in the output?

Ah, sorry. I didn't notice this. It's my fault.
Then this PR shouldn't be here. I'm sorry for wasting your time.