Wojciech Fibak: 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
83.6%z 4.31 - 2
82.3%z 4.21 - 3
78.9%z 4.00 - 4
78.5%z 4.00 - 5
74.1%z 3.95 - 6
69.4%z 3.90 - 7
66.1%z 3.86 - 8
60.8%z 3.81 - 9
60.1%z 3.80 - 10
55.8%z 3.75 - 11
51.6%z 3.71 - 12
47.4%z 3.63 - 13
47%z 3.60 - 14
46.4%z 3.56 - 15
46%z 3.54 - 16
45.1%z 3.48 - 17
44.7%z 3.45 - 18
44.6%z 3.45 - 19
44.5%z 3.44 - 20
44.5%z 3.44 - 21
44.2%z 3.42 - 22
44%z 3.41 - 23
43.5%z 3.38 - 24
43.5%z 3.37 - 25
43.4%z 3.37 - 26
43%z 3.34 - 27
42.7%z 3.33 - 28
41.6%z 3.25 - 29
41.6%z 3.25 - 30
41.5%z 3.24 - 31
41.3%z 3.23 - 32
41.3%z 3.23 - 33
41.2%z 3.23 - 34
41.2%z 3.22 - 35
41%z 3.21 - 36
40.8%z 3.20 - 37
40.7%z 3.20 - 38
40.4%z 3.18 - 39
40.1%z 3.16 - 40
40%z 3.15 - 41
39.9%z 3.14 - 42
39.8%z 3.13 - 43
39.7%z 3.13 - 44
39.5%z 3.12 - 45
39.4%z 3.11 - 46
39%z 3.09 - 47
39%z 3.08 - 48
38.7%z 3.06 - 49
38.6%z 3.06