Image Classification
Transformers
Safetensors
swinv2
deepfake-detection
ai-image-detection
dfd-arena
bitmind
Instructions to use pixelprism-ai/dfd-arena-mini with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use pixelprism-ai/dfd-arena-mini with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="pixelprism-ai/dfd-arena-mini") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")# Load model directly from transformers import AutoImageProcessor, AutoModelForImageClassification processor = AutoImageProcessor.from_pretrained("pixelprism-ai/dfd-arena-mini") model = AutoModelForImageClassification.from_pretrained("pixelprism-ai/dfd-arena-mini", device_map="auto") - Notebooks
- Google Colab
- Kaggle
Upload pixelprism_config.yaml with huggingface_hub
Browse files- pixelprism_config.yaml +12 -0
pixelprism_config.yaml
ADDED
|
@@ -0,0 +1,12 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# PixelPrism v0.1 — DFD Arena submission config
|
| 2 |
+
|
| 3 |
+
# Hugging Face repo where this detector's weights + processor live.
|
| 4 |
+
hf_repo: 'pixelprism-ai/dfd-arena-mini'
|
| 5 |
+
|
| 6 |
+
# Original upstream model (re-hosted under our repo for arena evaluator
|
| 7 |
+
# convenience; weights are byte-identical, license MIT preserved).
|
| 8 |
+
backbone_repo: 'haywoodsloan/ai-image-detector-deploy'
|
| 9 |
+
|
| 10 |
+
# Output class index that corresponds to "AI-generated" in the Swin V2 head.
|
| 11 |
+
# Verified at deployment: idx 0 = AI, idx 1 = real.
|
| 12 |
+
ai_label_idx: 0
|