Mark Takai: 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
89.1%z 4.76 - 2
81.9%z 4.17 - 3
78.2%z 4.00 - 4
58.8%z 3.78 - 5
56.7%z 3.76 - 6
55.5%z 3.75 - 7
47.3%z 3.62 - 8
47.1%z 3.61 - 9
46.2%z 3.55 - 10
46.2%z 3.55 - 11
46.1%z 3.54 - 12
46%z 3.53 - 13
45.6%z 3.51 - 14
44.9%z 3.47 - 15
44.7%z 3.46 - 16
44.7%z 3.45 - 17
44.5%z 3.44 - 18
43.9%z 3.40 - 19
43.9%z 3.40 - 20
43.6%z 3.38 - 21
42.9%z 3.34 - 22
42.2%z 3.29 - 23
42.2%z 3.29 - 24
41.6%z 3.25 - 25
41.5%z 3.25 - 26
41.3%z 3.23 - 27
41.3%z 3.23 - 28
40.8%z 3.20 - 29
40.6%z 3.19 - 30
40.4%z 3.17 - 31
39.8%z 3.13 - 32
39.6%z 3.12 - 33
39.2%z 3.10 - 34
39.1%z 3.09 - 35
39.1%z 3.09 - 36
38.9%z 3.08 - 37
38.8%z 3.07 - 38
38.8%z 3.07 - 39
38.5%z 3.05 - 40
38.3%z 3.04 - 41
38.2%z 3.03 - 42
38.2%z 3.03 - 43
38.2%z 3.03 - 44
38.1%z 3.03 - 45
38.1%z 3.02 - 46
37.9%z 3.01 - 47
37.6%z 2.99 - 48
37.6%z 2.99 - 49
37%z 2.96 - 50
37%z 2.95