What, exactly, is etaPos, and why often so large?

[gordonwatts]

It can often be around 2 or 3 (or even 4) different from the jet eta. Understand how we are doing the matching and make sure it is correct. This might be more of an issue with geometry and quite correct.

[gordonwatts]

Same for phiDir - and make sure to update the test in test_convert_divert which checks these to make sure they are reasonable!

[gordonwatts]

Here is an example of etaDir - most of them seem tob e zeros:

[gordonwatts]

There are two kinds of variables, according to the note:

• etaPos, phiPos - is the $\eta, \phi$ position of the muon segment - where it can be found in the detector w.r.t. the origin.
• etaDir, phiDir - is the direction the muon segment is pointing on.

I am not sure I see how these muon segments can be used with out detailed knowledge of how the detector works!

[gordonwatts]

Inconsistent phi/eta adjustment:

    mseg_deta_pos = mseg_list.etaPos - jet_list.eta  # type: ignore
    mseg_list["etaPos"] = mseg_deta_pos  # type: ignore
    mseg_dphi_pos = rectify(np, mseg_list.phiPos - jet_list.phi)  # type: ignore
    mseg_list["phiPos"] = mseg_dphi_pos  # type: ignore
    mseg_dphi_dir = rectify(np, mseg_list.phiDir - jet_list.phi)  # type: ignore
    mseg_list["phiDir"] = mseg_dphi_dir  # type: ignore

• Phi and eta of the position
• Phi of the direction

Why not the eta? :-)

[gordonwatts]

And here is the C++ pre-processing:

        for (int i=0; i < 30; i++){
                indexArray[i] = i;
                if (mseg_etaPos_arr[i] != 0)  mseg_etaPos_arr_copy[i] = mseg_etaPos_arr[i] - jet_eta;
                else mseg_etaPos_arr_copy[i] = mseg_etaPos_arr[i];
                if ( mseg_phiPos_arr[i] != 0 ) mseg_phiPos_arr[i] = mseg_phiPos_arr[i] - jet_phi;
                else mseg_phiPos_arr[i] = 9999999;
                mseg_etaDir_arr_copy[i] = mseg_etaDir_arr[i];
                if ( mseg_phiDir_arr[i] != 0) mseg_phiDir_arr_copy[i] = mseg_phiDir_arr[i] - jet_phi;
                else mseg_phiDir_arr_copy[i] = mseg_phiDir_arr[i];
                mseg_t0_arr_copy[i] = mseg_t0_arr[i];
                mseg_chiSquared_arr_copy[i] = mseg_chiSquared_arr[i];
        }

[gordonwatts]

Ok - so that preprocessing looks good! This all makes sense.