Miho Kanno: 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
99%z 6.70 - 2
92.9%z 5.39 - 3
91.7%z 5.16 - 4
91.5%z 5.12 - 5
90.5%z 4.92 - 6
87.2%z 4.60 - 7
86.8%z 4.57 - 8
86.4%z 4.54 - 9
86%z 4.51 - 10
85.6%z 4.47 - 11
84.6%z 4.39 - 12
84.5%z 4.39 - 13
84.1%z 4.35 - 14
83.4%z 4.29 - 15
83.1%z 4.27 - 16
81.8%z 4.16 - 17
81.5%z 4.14 - 18
81.4%z 4.13 - 19
81.2%z 4.11 - 20
80.6%z 4.07 - 21
80.2%z 4.03 - 22
75.7%z 3.97 - 23
74.8%z 3.96 - 24
73.8%z 3.95 - 25
72.7%z 3.94 - 26
69.1%z 3.90 - 27
68.9%z 3.90 - 28
68.9%z 3.89 - 29
65.8%z 3.86 - 30
64.9%z 3.85 - 31
64.3%z 3.85 - 32
63.5%z 3.84 - 33
59.5%z 3.79 - 34
59.4%z 3.79 - 35
55.3%z 3.75 - 36
54.3%z 3.74 - 37
52.7%z 3.72 - 38
52.1%z 3.71 - 39
52%z 3.71 - 40
51.6%z 3.71 - 41
51.3%z 3.70 - 42
50.1%z 3.69 - 43
48.9%z 3.68 - 44
47.6%z 3.64 - 45
47.4%z 3.63 - 46
47.2%z 3.62 - 47
47.2%z 3.61 - 48
47.1%z 3.61 - 49
47%z 3.60 - 50
46.5%z 3.57