Marco Silva: 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
91.4%z 5.09 - 2
62.2%z 3.82 - 3
54%z 3.73 - 4
48%z 3.67 - 5
47.3%z 3.62 - 6
47%z 3.60 - 7
46.6%z 3.57 - 8
46.1%z 3.54 - 9
45.9%z 3.53 - 10
45.5%z 3.51 - 11
45.2%z 3.49 - 12
45.1%z 3.48 - 13
45%z 3.47 - 14
44.1%z 3.42 - 15
44%z 3.41 - 16
44%z 3.41 - 17
43.7%z 3.39 - 18
43.6%z 3.38 - 19
43.5%z 3.37 - 20
43.4%z 3.37 - 21
43.1%z 3.35 - 22
43%z 3.34 - 23
42.9%z 3.34 - 24
42.5%z 3.31 - 25
42.2%z 3.29 - 26
42.2%z 3.29 - 27
42.1%z 3.29 - 28
42%z 3.28 - 29
41.9%z 3.27 - 30
41.8%z 3.27 - 31
41.7%z 3.26 - 32
41.3%z 3.24 - 33
41.2%z 3.22 - 34
41.1%z 3.22 - 35
40.9%z 3.21 - 36
40.8%z 3.20 - 37
40.7%z 3.19 - 38
40.6%z 3.19 - 39
40.6%z 3.19 - 40
40.5%z 3.18 - 41
40.1%z 3.16 - 42
40.1%z 3.15 - 43
40%z 3.15 - 44
39.7%z 3.13 - 45
39.6%z 3.12 - 46
39.6%z 3.12 - 47
39.4%z 3.11 - 48
39.3%z 3.11 - 49
39.3%z 3.10 - 50
39.2%z 3.10