Fernando Di Leo: 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
76.2%z 3.97 - 2
56.9%z 3.76 - 3
51.6%z 3.71 - 4
47.8%z 3.65 - 5
47.5%z 3.63 - 6
47%z 3.60 - 7
46.9%z 3.60 - 8
46.8%z 3.59 - 9
46.8%z 3.59 - 10
46.4%z 3.56 - 11
46.4%z 3.56 - 12
45.6%z 3.51 - 13
45.2%z 3.49 - 14
44.7%z 3.45 - 15
44.7%z 3.45 - 16
44.7%z 3.45 - 17
44.2%z 3.42 - 18
43.9%z 3.40 - 19
43.6%z 3.38 - 20
43.6%z 3.38 - 21
43.3%z 3.36 - 22
43.3%z 3.36 - 23
43.1%z 3.35 - 24
43%z 3.34 - 25
42.4%z 3.31 - 26
42.4%z 3.31 - 27
42.1%z 3.28 - 28
41.9%z 3.27 - 29
41.8%z 3.27 - 30
41.6%z 3.25 - 31
41.3%z 3.23 - 32
41.2%z 3.23 - 33
41.1%z 3.22 - 34
40.6%z 3.19 - 35
40.3%z 3.17 - 36
40%z 3.15 - 37
40%z 3.15 - 38
40%z 3.15 - 39
39.7%z 3.13 - 40
39.4%z 3.11 - 41
39.4%z 3.11 - 42
39.3%z 3.10 - 43
39.2%z 3.10 - 44
38.9%z 3.08 - 45
38.9%z 3.08 - 46
38.8%z 3.07 - 47
38.8%z 3.07 - 48
38.6%z 3.06 - 49
38.5%z 3.05 - 50
38.4%z 3.04