Update README.md
Browse files
README.md
CHANGED
|
@@ -10,8 +10,6 @@ library_name: transformers
|
|
| 10 |
|
| 11 |
# AI and Human Image Classification Model - v1
|
| 12 |
A fine-tuned model trained on 60,000 AI-generated and 60,000 human images.
|
| 13 |
-
The model demonstrates strong capability in detecting high-quality, state-of-the-art AI-generated images from models such as Midjourney v6.1, Flux 1.1 Pro, Stable Diffusion 3.5, GPT-4o, and other trending generation models.
|
| 14 |
-
|
| 15 |
Detailed Training Code is available here: [blog/ai/fine-tuning-siglip2](https://exnrt.com/blog/ai/fine-tuning-siglip2/)
|
| 16 |
|
| 17 |
## Evaluation Metrics
|
|
@@ -32,7 +30,10 @@ Detailed Training Code is available here: [blog/ai/fine-tuning-siglip2](https://
|
|
| 32 |
- **Eval Loss:** 0.0551
|
| 33 |
- **Eval Runtime:** 0:02:35.78
|
| 34 |
- **Eval Samples/Sec:** 212.533
|
| 35 |
-
- **Eval Steps/Sec:** 6.644
|
|
|
|
|
|
|
|
|
|
| 36 |
|
| 37 |
### 🔦 Prediction Metrics (on test set):
|
| 38 |
|
|
|
|
| 10 |
|
| 11 |
# AI and Human Image Classification Model - v1
|
| 12 |
A fine-tuned model trained on 60,000 AI-generated and 60,000 human images.
|
|
|
|
|
|
|
| 13 |
Detailed Training Code is available here: [blog/ai/fine-tuning-siglip2](https://exnrt.com/blog/ai/fine-tuning-siglip2/)
|
| 14 |
|
| 15 |
## Evaluation Metrics
|
|
|
|
| 30 |
- **Eval Loss:** 0.0551
|
| 31 |
- **Eval Runtime:** 0:02:35.78
|
| 32 |
- **Eval Samples/Sec:** 212.533
|
| 33 |
+
- **Eval Steps/Sec:** 6.644
|
| 34 |
+
|
| 35 |
+
### NOTE: Users report
|
| 36 |
+
Some users reported overfitting issues
|
| 37 |
|
| 38 |
### 🔦 Prediction Metrics (on test set):
|
| 39 |
|