Philip Bradbourn: 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
69.5%z 3.90 - 2
48.5%z 3.67 - 3
47.7%z 3.64 - 4
47.4%z 3.63 - 5
46.9%z 3.59 - 6
45.8%z 3.52 - 7
44.9%z 3.46 - 8
44.1%z 3.42 - 9
43.6%z 3.38 - 10
43.4%z 3.37 - 11
43.2%z 3.35 - 12
42.7%z 3.33 - 13
42.5%z 3.31 - 14
41.8%z 3.26 - 15
41.5%z 3.24 - 16
40.8%z 3.20 - 17
40.4%z 3.17 - 18
40%z 3.15 - 19
39.7%z 3.13 - 20
39.5%z 3.12 - 21
39.4%z 3.11 - 22
39.3%z 3.10 - 23
38.6%z 3.06 - 24
38.6%z 3.06 - 25
38.4%z 3.04 - 26
38.3%z 3.04 - 27
38.2%z 3.03 - 28
38.1%z 3.02 - 29
38%z 3.02 - 30
37.6%z 2.99 - 31
37.5%z 2.98 - 32
37.4%z 2.98 - 33
37.3%z 2.98 - 34
37.2%z 2.97 - 35
37.1%z 2.96 - 36
37%z 2.95 - 37
36.8%z 2.94 - 38
36.7%z 2.94 - 39
36.5%z 2.92 - 40
36.5%z 2.92 - 41
36.5%z 2.92 - 42
36.4%z 2.92 - 43
36.4%z 2.91 - 44
36.4%z 2.91 - 45
36.2%z 2.90 - 46
36.2%z 2.90 - 47
36.2%z 2.90 - 48
36.1%z 2.89 - 49
36%z 2.89 - 50
36%z 2.89