Closed yipengsun closed 1 year ago
Here's a comparison of updated vs old efficiencies. Note that the legends are wrong. Both black and red are the same data, the only difference is ProbNNk ghost effficiency (and updated UBDT efficiency, which doesn't do much, as can be seen in the ISO and 1OS skims).
I think you should be careful on your first statement here.
Surprisingly, most of the first isolation tracks are actually ghost. Looking at MD of 11574021 (Dst0Mu)
you're looking at a sample where there is no true B daughter for the isolation to find. So not having MC matches just tells you that the isolation is very good at not getting distracted by PV pions and Kaons. I think you'd find differently if you look at truly anti-isolated MC (DD, Dstst, etc)
As Phoebe suggested, the 2OS problem might be: Some of the isolation tracks fall out of binning range, and due to the way the run 2 histos are constructed, under/overflow bins currently are garbage:
## z ∈ [-inf, 0.00)
╒════════════╤═════════════════╤═════════════════╤═════════════════╤═════════════════╕
│ x / y │ -inf │ 1.70 │ 3.60 │ 5.00 │
│ │ 1.70 │ 3.60 │ 5.00 │ inf │
╞════════════╪═════════════════╪═════════════════╪═════════════════╪═════════════════╡
│ -inf │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 3,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 3,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 6,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 6,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 9,300.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 9,300.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 10,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 10,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 15,600.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 15,600.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 20,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 20,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 40,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 40,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 100,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼─────────────────┼─────────────────┤
│ 100,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ inf │ │ │ │ │
╘════════════╧═════════════════╧═════════════════╧═════════════════╧═════════════════╛
## z ∈ [0.00, 150.00)
╒════════════╤═════════════════╤══════════════════╤══════════════════╤══════════════════╕
│ x / y │ -inf │ 1.70 │ 3.60 │ 5.00 │
│ │ 1.70 │ 3.60 │ 5.00 │ inf │
╞════════════╪═════════════════╪══════════════════╪══════════════════╪══════════════════╡
│ -inf │ 0.0000 ± 0.0000 │ 20.0000 ± 4.4721 │ 20.0000 ± 4.4721 │ 0.0000 ± 0.0000 │
│ 3,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 3,000.00 │ 1.0000 ± 1.0000 │ 0.2050 ± 0.0065 │ 0.0581 ± 0.0020 │ 0.0000 ± 0.0000 │
│ 6,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 6,000.00 │ 0.0000 ± 0.0000 │ 0.3975 ± 0.0141 │ 0.1142 ± 0.0037 │ 33.0000 ± 5.7446 │
│ 9,300.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 9,300.00 │ 0.0000 ± 0.0000 │ 0.3311 ± 0.0425 │ 0.2451 ± 0.0138 │ 15.0000 ± 3.8730 │
│ 10,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 10,000.00 │ 0.0000 ± 0.0000 │ 0.3315 ± 0.0184 │ 0.2556 ± 0.0065 │ 64.0000 ± 8.0000 │
│ 15,600.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 15,600.00 │ 0.0000 ± 0.0000 │ 0.2383 ± 0.0325 │ 0.2760 ± 0.0118 │ 24.0000 ± 4.8990 │
│ 20,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 20,000.00 │ 0.0000 ± 0.0000 │ 0.2908 ± 0.0294 │ 0.2770 ± 0.0098 │ 22.0000 ± 4.6904 │
│ 40,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 40,000.00 │ 0.0000 ± 0.0000 │ 0.3168 ± 0.0527 │ 0.2509 ± 0.0159 │ 7.0000 ± 2.6458 │
│ 100,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 100,000.00 │ 0.0000 ± 0.0000 │ 6.0000 ± 2.4495 │ 38.0000 ± 6.1644 │ 1.0000 ± 1.0000 │
│ inf │ │ │ │ │
╘════════════╧═════════════════╧══════════════════╧══════════════════╧══════════════════╛
## z ∈ [150.00, 225.00)
╒════════════╤═════════════════╤══════════════════╤══════════════════╤══════════════════╕
│ x / y │ -inf │ 1.70 │ 3.60 │ 5.00 │
│ │ 1.70 │ 3.60 │ 5.00 │ inf │
╞════════════╪═════════════════╪══════════════════╪══════════════════╪══════════════════╡
│ -inf │ 0.0000 ± 0.0000 │ 12.0000 ± 3.4641 │ 10.0000 ± 3.1623 │ 1.0000 ± 1.0000 │
│ 3,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 3,000.00 │ 0.0000 ± 0.0000 │ 0.1501 ± 0.0054 │ 0.0430 ± 0.0016 │ 0.0000 ± 0.0000 │
│ 6,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 6,000.00 │ 0.0000 ± 0.0000 │ 0.3205 ± 0.0130 │ 0.0876 ± 0.0032 │ 31.0000 ± 5.5678 │
│ 9,300.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 9,300.00 │ 0.0000 ± 0.0000 │ 0.4492 ± 0.0390 │ 0.1822 ± 0.0124 │ 9.0000 ± 3.0000 │
│ 10,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 10,000.00 │ 0.0000 ± 0.0000 │ 0.3750 ± 0.0195 │ 0.1990 ± 0.0057 │ 63.0000 ± 7.9373 │
│ 15,600.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 15,600.00 │ 0.0000 ± 0.0000 │ 0.2414 ± 0.0366 │ 0.2368 ± 0.0108 │ 31.0000 ± 5.5678 │
│ 20,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 20,000.00 │ 0.0000 ± 0.0000 │ 0.2500 ± 0.0312 │ 0.2723 ± 0.0095 │ 33.0000 ± 5.7446 │
│ 40,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 40,000.00 │ 0.0000 ± 0.0000 │ 0.2778 ± 0.0625 │ 0.2573 ± 0.0160 │ 7.0000 ± 2.6458 │
│ 100,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 100,000.00 │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 36.0000 ± 6.0000 │ 1.0000 ± 1.0000 │
│ inf │ │ │ │ │
╘════════════╧═════════════════╧══════════════════╧══════════════════╧══════════════════╛
## z ∈ [225.00, 600.00)
╒════════════╤═════════════════╤══════════════════╤══════════════════╤══════════════════╕
│ x / y │ -inf │ 1.70 │ 3.60 │ 5.00 │
│ │ 1.70 │ 3.60 │ 5.00 │ inf │
╞════════════╪═════════════════╪══════════════════╪══════════════════╪══════════════════╡
│ -inf │ 0.0000 ± 0.0000 │ 14.0000 ± 3.7417 │ 6.0000 ± 2.4495 │ 0.0000 ± 0.0000 │
│ 3,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 3,000.00 │ 0.0000 ± 0.0000 │ 0.1114 ± 0.0045 │ 0.0404 ± 0.0015 │ 0.0000 ± 0.0000 │
│ 6,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 6,000.00 │ 0.0000 ± 0.0000 │ 0.2770 ± 0.0119 │ 0.0597 ± 0.0026 │ 10.0000 ± 3.1623 │
│ 9,300.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 9,300.00 │ 0.0000 ± 0.0000 │ 0.3476 ± 0.0357 │ 0.1207 ± 0.0099 │ 8.0000 ± 2.8284 │
│ 10,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 10,000.00 │ 0.0000 ± 0.0000 │ 0.3338 ± 0.0185 │ 0.1370 ± 0.0047 │ 44.0000 ± 6.6332 │
│ 15,600.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 15,600.00 │ 0.0000 ± 0.0000 │ 0.2655 ± 0.0372 │ 0.1935 ± 0.0098 │ 21.0000 ± 4.5826 │
│ 20,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 20,000.00 │ 0.0000 ± 0.0000 │ 0.3132 ± 0.0379 │ 0.2324 ± 0.0091 │ 36.0000 ± 6.0000 │
│ 40,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 40,000.00 │ 0.0000 ± 0.0000 │ 0.2778 ± 0.0739 │ 0.2778 ± 0.0174 │ 14.0000 ± 3.7417 │
│ 100,000.00 │ │ │ │ │
├────────────┼─────────────────┼──────────────────┼──────────────────┼──────────────────┤
│ 100,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 24.0000 ± 4.8990 │ 0.0000 ± 0.0000 │
│ inf │ │ │ │ │
╘════════════╧═════════════════╧══════════════════╧══════════════════╧══════════════════╛
## z ∈ [600.00, inf)
╒════════════╤═════════════════╤═════════════════╤══════════════════╤═════════════════╕
│ x / y │ -inf │ 1.70 │ 3.60 │ 5.00 │
│ │ 1.70 │ 3.60 │ 5.00 │ inf │
╞════════════╪═════════════════╪═════════════════╪══════════════════╪═════════════════╡
│ -inf │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ 3,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 3,000.00 │ 0.0000 ± 0.0000 │ 6.0000 ± 2.4495 │ 17.0000 ± 4.1231 │ 0.0000 ± 0.0000 │
│ 6,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 6,000.00 │ 0.0000 ± 0.0000 │ 3.0000 ± 1.7321 │ 4.0000 ± 2.0000 │ 0.0000 ± 0.0000 │
│ 9,300.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 9,300.00 │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 1.0000 ± 1.0000 │ 0.0000 ± 0.0000 │
│ 10,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 10,000.00 │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 3.0000 ± 1.7321 │ 0.0000 ± 0.0000 │
│ 15,600.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 15,600.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 0.0000 ± 0.0000 │
│ 20,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 20,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 5.0000 ± 2.2361 │ 0.0000 ± 0.0000 │
│ 40,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 40,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 1.0000 ± 1.0000 │ 0.0000 ± 0.0000 │
│ 100,000.00 │ │ │ │ │
├────────────┼─────────────────┼─────────────────┼──────────────────┼─────────────────┤
│ 100,000.00 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │ 0.0000 ± 0.0000 │
│ inf │ │ │ │ │
╘════════════╧═════════════════╧═════════════════╧══════════════════╧═════════════════╛
I should make under/overflow bins meaningful and try again.
Our current MC sample doesn't cover P > 100 GeV
region. I can include a guard to make P > 100 GeV
events use nearby bins.
Efficiency updated with sensible upper/under (by sensible, I mean between [0, 1]). I'll rebuild step-2 ntuples and fit templates.
There was also a problem for 1OS weight: The weight has an incorrect type of bool
, instead of double
. I've fixed that and regenerated templates.
All 3 fit steps converged with updated ProbNNk weights.
Surprisingly, most of the first isolation tracks are actually ghost. Looking at MD of
11574021
(Dst0Mu
):These efficiencies are used in:
https://github.com/umd-lhcb/lhcb-ntuples-gen/blob/e35ae99ec16e889c2a709b3bd01c199a50c84103/postprocess/rdx-run2/rdx-run2_oldcut.yml#L674-L694
I tried to update the True ghost ProbNNk efficiency using run 2 normalization MC, but the efficiencies are very different from that of run 1.
run 1
File located at:
https://github.com/umd-lhcb/lhcb-ntuples-gen/blob/master/run2-rdx/reweight/pid/root-run1-rdx_iso_oldcut/run1-rdx-2012-md-ghost_nnk_gt-p_eta_ntracks.root
With the following efficiencies (
kISOeff
)run 2
File located at: https://github.com/umd-lhcb/lhcb-ntuples-gen/blob/master/run2-rdx/reweight/pid/root-run2-rdx_iso_oldcut_ghost/run2-rdx-2016-md-ghost_nnk_gt-p_eta_ntracks.root
With efficiencies (
eff
):