Linda Purl: 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
91.6%z 5.14 - 2
90.2%z 4.86 - 3
87.4%z 4.62 - 4
85.4%z 4.46 - 5
84.8%z 4.41 - 6
84.3%z 4.36 - 7
84.2%z 4.36 - 8
83.7%z 4.32 - 9
83.6%z 4.31 - 10
82.4%z 4.21 - 11
81.7%z 4.15 - 12
81.3%z 4.13 - 13
81.1%z 4.11 - 14
78.9%z 4.00 - 15
78.3%z 4.00 - 16
77.3%z 3.99 - 17
77%z 3.98 - 18
73.9%z 3.95 - 19
73.6%z 3.95 - 20
72.1%z 3.93 - 21
71.9%z 3.93 - 22
71.1%z 3.92 - 23
70.4%z 3.91 - 24
70.1%z 3.91 - 25
68.3%z 3.89 - 26
67.1%z 3.88 - 27
66.3%z 3.87 - 28
64.3%z 3.85 - 29
63.3%z 3.83 - 30
63.2%z 3.83 - 31
61.4%z 3.81 - 32
60.5%z 3.80 - 33
60.2%z 3.80 - 34
57.1%z 3.77 - 35
55.9%z 3.75 - 36
55.7%z 3.75 - 37
54.3%z 3.74 - 38
54.3%z 3.74 - 39
52.9%z 3.72 - 40
48.8%z 3.67 - 41
48%z 3.67 - 42
47.9%z 3.66 - 43
47.6%z 3.64 - 44
47.5%z 3.64 - 45
47.4%z 3.63 - 46
47.1%z 3.61 - 47
46.4%z 3.57 - 48
46.3%z 3.56 - 49
46.3%z 3.55 - 50
46.2%z 3.55