Tomasz Kot: 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
85%z 4.43 - 2
84.1%z 4.35 - 3
68%z 3.89 - 4
65%z 3.85 - 5
61.8%z 3.82 - 6
54.1%z 3.73 - 7
54%z 3.73 - 8
49.2%z 3.68 - 9
47.7%z 3.65 - 10
46.6%z 3.58 - 11
46.4%z 3.56 - 12
46.1%z 3.54 - 13
45.6%z 3.51 - 14
45.2%z 3.49 - 15
44.9%z 3.47 - 16
44.9%z 3.46 - 17
44.7%z 3.45 - 18
44.1%z 3.42 - 19
44.1%z 3.41 - 20
43.7%z 3.39 - 21
43.5%z 3.37 - 22
43.4%z 3.37 - 23
43.2%z 3.36 - 24
43%z 3.34 - 25
42.9%z 3.34 - 26
42.9%z 3.33 - 27
42.8%z 3.33 - 28
42.7%z 3.33 - 29
42.3%z 3.30 - 30
42.3%z 3.30 - 31
41.6%z 3.25 - 32
41.4%z 3.24 - 33
41.4%z 3.24 - 34
41.1%z 3.22 - 35
40.8%z 3.20 - 36
40.4%z 3.18 - 37
40.1%z 3.16 - 38
40.1%z 3.16 - 39
39.8%z 3.14 - 40
39.7%z 3.13 - 41
39.7%z 3.13 - 42
39.5%z 3.12 - 43
39.2%z 3.10 - 44
39%z 3.08 - 45
39%z 3.08 - 46
38.9%z 3.08 - 47
38.8%z 3.07 - 48
38.7%z 3.07 - 49
38.6%z 3.06 - 50
38.4%z 3.04