Downloads · 30 days
10
2% of all-time downloads
deepfake-image-detector-new-latest
shivani1511/deepfake-image-detector-new-latest
deepfake-image-detector-new-latest is a image classification model from shivani1511. Use it when you need a label for an image. The card lists the license as mit.
- Test Accuracy: 73.17% - Best Validation Accuracy: 86.27% - Best Epoch: 11 - Planned Epochs: 12 - Actual Epochs Trained: 12 (early stopping applied)
All-time downloads
585
Public
Parameters
85.8M
6 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.pth5.6 GB · 94%
At a glance
- Task
- Image Classification
- License
- mit
- Model type
- vit
- Access
- Public
- Created
- Jun 27, 2025
- Updated
- Jun 27, 2025
- SHA
- be3c0d62
Model card
From the Hugging Face model README
Deepfake Image Detector (CILab Fine-Tuned)
Model Performance
- Test Accuracy: 73.17%
- Best Validation Accuracy: 86.27%
- Best Epoch: 11
- Planned Epochs: 12
- Actual Epochs Trained: 12 (early stopping applied)
Dataset
- Training: 1,632 images
- Validation: 204 images
- Test: 205 images
Training Details
Training was stopped early at epoch 12 due to early stopping criteria being met. The best model was achieved at epoch 11 with validation accuracy of 86.27%.
Usage
from transformers import ViTForImageClassification, ViTFeatureExtractor
model = ViTForImageClassification.from_pretrained('shivani1511/deepfake-image-detector-new-latest')
feature_extractor = ViTFeatureExtractor.from_pretrained('shivani1511/deepfake-image-detector-new-latest')
Notes
- Fine-tuned on CILab dataset to address AI-generated fake detection (e.g., StyleGAN).
- Base model: shivani1511/deepfake-image-detector-new (Vision Transformer).
- Improvements: Enhanced data augmentation, class-weighted loss, Mixup, more unfrozen layers.
- Task
- Image Classification
- Type
- vit
- License
- mit
- Created
- Jun 27, 2025
- Updated
- Jun 27, 2025