YoussefAboelwafa/Face-Age_UTKFACE_FGNET
Face-Age_UTKFACE_FGNET Dataset Overview Property Value Total Images 32,173 Age Range 0 - 90 years Age Classes 91 (folders 0-90) Mean Image Dimensions 269 x 271 px Sources UTKFace dataset, FGNET dataset HuggingFace Repo YoussefAboelwafa/Face-Age_UTKFACE_FGNET Structure dataset/ ├── train/ │ ├── 0/ # Age 0 (34 images) │ ├── 1/ # Age 1 (1,039 images) │ │ ... │ └── 90/ # Age 90 (82 images)… See the full description on the dataset page: https://huggingface.co/datasets/YoussefAboelwafa/Face-Age_UTKFACE_FGNET.
0907
Face-AgeUTKFACEFGNET Dataset
Overview
Structure
dataset/
├── train/
│ ├── 0/ # Age 0 (34 images)
│ ├── 1/ # Age 1 (1,039 images)
│ │ ...
│ └── 90/ # Age 90 (82 images)
└── val/
├── 0/ # Age 0 (9 images)
├── 1/ # Age 1 (260 images)
│ ...
└── 90/ # Age 90 (21 images)Each folder contains face images labeled by age (integer 0-90).
Train/Val Split
- Split ratio: 80:20 (train:val)
- Random seed: 42 (reproducible)
- Method: Stratified per age class — each age folder split 80/20 independently
Age Distribution
Age Count
0 43 █
1 1,299 ████████████████████████████████
2 603 ██████████████
3 352 ████████
4 416 ██████████
5 323 ███████
6 236 █████
7 243 █████
8 338 ████████
9 225 █████
10 239 █████
11 149 ███
12 223 █████
13 192 ████
14 314 ███████
15 335 ████████
16 475 ███████████
17 349 ████████
18 502 ████████████
19 301 ███████
20 482 ███████████
21 505 ████████████
22 589 ██████████████
23 659 ████████████████
24 1,079 ██████████████████████████
25 1,066 ██████████████████████████
26 2,434 ████████████████████████████████████████████████████████████
27 855 █████████████████████
28 1,147 ████████████████████████████
29 743 ██████████████████
30 1,060 ██████████████████████████
31 485 ███████████
32 842 ████████████████████
33 296 ███████
34 573 ██████████████
35 1,038 █████████████████████████
36 613 ███████████████
37 440 ██████████
38 454 ███████████
39 379 █████████
40 702 █████████████████
41 236 █████
42 370 █████████
43 244 ██████
44 200 ████
45 535 █████████████
46 230 █████
47 265 ██████
48 222 █████
49 225 █████
50 539 █████████████
51 211 █████
52 311 ███████
53 288 ███████
54 399 █████████
55 321 ███████
56 277 ██████
57 134 ███
58 326 ████████
59 123 ███
60 369 █████████
61 192 ████
62 154 ███
63 124 ███
64 73 █
65 295 ███████
66 95 ██
67 113 ██
68 115 ██
69 75 █
70 197 ████
71 43 █
72 111 ██
73 73 █
74 38 █
75 165 ████
76 65 █
77 36 █
78 72 █
79 29 █
80 164 ████
81 25 █
82 45 █
83 21 █
84 32 █
85 167 ████
86 38 █
87 16 █
88 37 █
89 37 █
90 103 ██Age Group Breakdown
Distribution Statistics
Analysis
Class Imbalance
The dataset exhibits significant class imbalance:
- Overrepresented: Ages 20-49 (young adults and adults) contain 58.9% of all images
- Peak concentration: Age 26 alone has 2,434 images (7.6% of dataset)
- Underrepresented: Ages 80-90 have only 685 images total (2.1%)
- Extreme scarcity: Age 87 has only 16 images
<hr>
