Hugging Face's logo

ash12321
/
deepfake-detector-v13

Model card Files
xet
Community
deepfake-detector-v13
File size: 2,247 Bytes 
c37abab
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
 
{
  "model_name": "DeepFake Detector V13",
  "version": "13.0",
  "architecture": "3-Model Ensemble",
  "total_parameters": "699M",
  "description": "Large-scale ensemble with ConvNeXt-Large (198M), ViT-Large (304M), and Swin-Large (197M)",
  "models": [
    {
      "id": 1,
      "name": "Model 13.1",
      "backbone": "convnext_large",
      "parameters": "198M",
      "dropout": 0.3,
      "batch_size": 32,
      "best_f1": 0.8971,
      "file": "model_1.safetensors"
    },
    {
      "id": 2,
      "name": "Model 13.2",
      "backbone": "vit_large_patch16_224",
      "parameters": "304M",
      "dropout": 0.35,
      "batch_size": 24,
      "best_f1": 0.9382,
      "file": "model_2.safetensors"
    },
    {
      "id": 3,
      "name": "Model 13.3",
      "backbone": "swin_large_patch4_window7_224",
      "parameters": "197M",
      "dropout": 0.3,
      "batch_size": 32,
      "best_f1": 0.9586,
      "file": "model_3.safetensors"
    }
  ],
  "ensemble_performance": {
    "average_f1": 0.9313,
    "best_individual_f1": 0.9586,
    "total_training_time_hours": 6.1
  },
  "training": {
    "epochs_per_model": 10,
    "learning_rates": [
      2e-05,
      1.5e-05,
      1.8e-05
    ],
    "weight_decay": 0.0003,
    "label_smoothing": 0.12,
    "gradient_accumulation": 4,
    "mixed_precision": true,
    "criterion": "FocalLossSmooth (alpha=0.25, gamma=2.5)",
    "optimizer": "AdamW",
    "scheduler": "CosineAnnealingWarmRestarts"
  },
  "preprocessing": {
    "image_size": 224,
    "normalization": {
      "mean": [
        0.485,
        0.456,
        0.406
      ],
      "std": [
        0.229,
        0.224,
        0.225
      ]
    },
    "augmentations": [
      "RandomHorizontalFlip(p=0.5)",
      "RandomRotation(degrees=12)",
      "ColorJitter(brightness=0.15, contrast=0.15, saturation=0.15)"
    ]
  },
  "inference": {
    "ensemble_method": "average",
    "threshold": 0.5,
    "description": "Average predictions from all 3 models for final classification"
  },
  "requirements": [
    "torch>=2.0.0",
    "timm>=0.9.0",
    "torchvision>=0.15.0",
    "numpy",
    "pillow",
    "safetensors"
  ],
  "dataset": "ash12321/deepfake-v13-dataset",
  "predecessor": "ash12321/deepfake-detector-v12"
}