PrHybridSeeding - Issues in DD4HEP builds
See discussion in #370 (comment 5875097)
PrHybridSeeding is giving very different results between Detdesc and dd4hep.
DetDesc
PrHybridSeeding INFO Number of counters : 21
| Counter | # | sum | mean/eff^* | rms/err^* | min | max |
| "Created T2x1 three-hit combinations in case 0" | 130424 | 78183 | 0.59945 | 0.60972 | 0.0000 | 6.0000 |
| "Created T2x1 three-hit combinations in case 1" | 201837 | 131353 | 0.65079 | 0.71665 | 0.0000 | 10.000 |
| "Created T2x1 three-hit combinations in case 2" | 381682 | 301339 | 0.78950 | 0.95432 | 0.0000 | 20.000 |
| "Created XZ tracks (part 0)" | 219 | 11325 | 51.712 | 47.114 | 1.0000 | 405.00 |
| "Created XZ tracks (part 1)" | 219 | 11147 | 50.900 | 38.520 | 4.0000 | 209.00 |
| "Created XZ tracks in case 0" | 146 | 7120 | 48.767 | 29.595 | 3.0000 | 158.00 |
| "Created XZ tracks in case 1" | 146 | 8434 | 57.767 | 42.520 | 6.0000 | 308.00 |
| "Created XZ tracks in case 2" | 146 | 6918 | 47.384 | 52.995 | 1.0000 | 405.00 |
| "Created full hit combinations in case 0" | 10758 | 10758 | 1.0000 | 0.0000 | 1.0000 | 1.0000 |
| "Created full hit combinations in case 1" | 9873 | 9873 | 1.0000 | 0.0000 | 1.0000 | 1.0000 |
| "Created full hit combinations in case 2" | 10648 | 10648 | 1.0000 | 0.0000 | 1.0000 | 1.0000 |
| "Created seed tracks" | 146 | 7732 | 52.959 | 20.279 | 15.000 | 101.00 |
| "Created seed tracks (part 0)" | 73 | 4438 | 60.795 | 24.073 | 15.000 | 115.00 |
| "Created seed tracks (part 1)" | 73 | 4239 | 58.068 | 22.743 | 17.000 | 123.00 |
| "Created seed tracks in case 0" | 146 | 3874 | 26.534 | 11.351 | 3.0000 | 58.000 |
| "Created seed tracks in case 1" | 146 | 7300 | 50.000 | 19.562 | 13.000 | 102.00 |
| "Created seed tracks in case 2" | 146 | 8222 | 56.315 | 22.243 | 15.000 | 114.00 |
| "Created seed tracks in recovery step" | 73 | 455 | 6.2329 | 4.1727 | 0.0000 | 20.000 |
| "Created two-hit combinations in case 0" | 25034 | 560295 | 22.381 | 13.668 | 0.0000 | 88.000 |
| "Created two-hit combinations in case 1" | 22752 | 735783 | 32.339 | 16.713 | 0.0000 | 120.00 |
| "Created two-hit combinations in case 2" | 19255 | 1000491 | 51.960 | 26.627 | 3.0000 | 166.00 |
DD4HEP
PrHybridSeeding INFO Number of counters : 15
| Counter | # | sum | mean/eff^* | rms/err^* | min | max |
| "Created XZ tracks (part 0)" | 219 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created XZ tracks (part 1)" | 219 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created XZ tracks in case 0" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created XZ tracks in case 1" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created XZ tracks in case 2" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks (part 0)" | 73 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks (part 1)" | 73 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks in case 0" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks in case 1" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks in case 2" | 146 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created seed tracks in recovery step" | 73 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created two-hit combinations in case 0" | 25034 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created two-hit combinations in case 1" | 25891 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
| "Created two-hit combinations in case 2" | 25034 | 0 | 0.0000 | 0.0000 | 0.0000 | 0.0000 |
Likely related to known issues with FT, but we need the maintainer of PrHybridSeeding to dig in to confirm what exactly is leading to the differences in the two builds.
@lohenry @decianm FYI Can you assign to the maintainer of this alg ?
Edited by Christopher Rob Jones