Instructions to use prithivMLmods/Fire-Detection-Siglip2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use prithivMLmods/Fire-Detection-Siglip2 with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("image-classification", model="prithivMLmods/Fire-Detection-Siglip2") pipe("https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/hub/parrots.png")# Load model directly from transformers import AutoProcessor, AutoModelForImageClassification processor = AutoProcessor.from_pretrained("prithivMLmods/Fire-Detection-Siglip2") model = AutoModelForImageClassification.from_pretrained("prithivMLmods/Fire-Detection-Siglip2") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -9,9 +9,24 @@ library_name: transformers
|
|
| 9 |
tags:
|
| 10 |
- fire-detection
|
| 11 |
---
|
|
|
|
|
|
|
| 12 |
# **Fire-Detection-Siglip2**
|
| 13 |
|
| 14 |
-
**Fire-Detection-Siglip2** is an image classification vision-language encoder model fine-tuned from google/siglip2-base-patch16-224 for a single-label classification task. It is designed to detect fire, smoke, or normal conditions using the SiglipForImageClassification architecture.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 15 |
|
| 16 |
The model categorizes images into three classes:
|
| 17 |
- **Class 0:** "Fire" – The image shows active fire.
|
|
@@ -66,17 +81,6 @@ if __name__ == "__main__":
|
|
| 66 |
iface.launch()
|
| 67 |
```
|
| 68 |
|
| 69 |
-
Classification report:
|
| 70 |
-
|
| 71 |
-
precision recall f1-score support
|
| 72 |
-
|
| 73 |
-
fire 0.9940 0.9881 0.9911 1010
|
| 74 |
-
normal 0.9892 0.9941 0.9916 1010
|
| 75 |
-
smoke 0.9990 1.0000 0.9995 1010
|
| 76 |
-
|
| 77 |
-
accuracy 0.9941 3030
|
| 78 |
-
macro avg 0.9941 0.9941 0.9941 3030
|
| 79 |
-
weighted avg 0.9941 0.9941 0.9941 3030
|
| 80 |
|
| 81 |
# **Intended Use:**
|
| 82 |
|
|
|
|
| 9 |
tags:
|
| 10 |
- fire-detection
|
| 11 |
---
|
| 12 |
+

|
| 13 |
+
|
| 14 |
# **Fire-Detection-Siglip2**
|
| 15 |
|
| 16 |
+
> **Fire-Detection-Siglip2** is an image classification vision-language encoder model fine-tuned from google/siglip2-base-patch16-224 for a single-label classification task. It is designed to detect fire, smoke, or normal conditions using the SiglipForImageClassification architecture.
|
| 17 |
+
|
| 18 |
+
Classification report:
|
| 19 |
+
|
| 20 |
+
precision recall f1-score support
|
| 21 |
+
|
| 22 |
+
fire 0.9940 0.9881 0.9911 1010
|
| 23 |
+
normal 0.9892 0.9941 0.9916 1010
|
| 24 |
+
smoke 0.9990 1.0000 0.9995 1010
|
| 25 |
+
|
| 26 |
+
accuracy 0.9941 3030
|
| 27 |
+
macro avg 0.9941 0.9941 0.9941 3030
|
| 28 |
+
weighted avg 0.9941 0.9941 0.9941 3030
|
| 29 |
+
|
| 30 |
|
| 31 |
The model categorizes images into three classes:
|
| 32 |
- **Class 0:** "Fire" – The image shows active fire.
|
|
|
|
| 81 |
iface.launch()
|
| 82 |
```
|
| 83 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 84 |
|
| 85 |
# **Intended Use:**
|
| 86 |
|