Marcus Haber: raw computed results
The raw ranking from the production similarity calculation (face-embedding centroid cosine → gender-cohort z-normalization). The web-opinion crown, curation, and fusion re-rank are removed.
- 1
84.3%z 4.37 - 2
81.5%z 4.14 - 3
81%z 4.10 - 4
80.1%z 4.03 - 5
69.3%z 3.90 - 6
56.1%z 3.76 - 7
50%z 3.69 - 8
46.5%z 3.57 - 9
46.5%z 3.57 - 10
46.1%z 3.54 - 11
45.7%z 3.52 - 12
45.6%z 3.51 - 13
44.6%z 3.45 - 14
44.2%z 3.42 - 15
43.5%z 3.37 - 16
43.2%z 3.36 - 17
43.1%z 3.35 - 18
43%z 3.35 - 19
43%z 3.34 - 20
42.9%z 3.33 - 21
42.6%z 3.32 - 22
41.9%z 3.27 - 23
41.5%z 3.25 - 24
41.5%z 3.24 - 25
41.2%z 3.23 - 26
41.1%z 3.22 - 27
40.8%z 3.20 - 28
40.8%z 3.20 - 29
40.7%z 3.19 - 30
40.5%z 3.18 - 31
40.4%z 3.17 - 32
40.2%z 3.16 - 33
40%z 3.15 - 34
39.9%z 3.14 - 35
39.9%z 3.14 - 36
39.8%z 3.14 - 37
39.8%z 3.13 - 38
39.1%z 3.09 - 39
38.9%z 3.08 - 40
38.9%z 3.08 - 41
38.6%z 3.06 - 42
38.5%z 3.05 - 43
38.4%z 3.04 - 44
38.3%z 3.04 - 45
38.3%z 3.04 - 46
38.3%z 3.04 - 47
37.9%z 3.01 - 48
37.9%z 3.01 - 49
37.8%z 3.01 - 50
37.5%z 2.99