Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,40 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: mit
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
base_model: google/vivit-b-16x2
|
| 4 |
+
tags:
|
| 5 |
+
- cctv-surveillance
|
| 6 |
+
- video-classification
|
| 7 |
+
metrics:
|
| 8 |
+
- accuracy
|
| 9 |
+
- f1
|
| 10 |
+
- recall
|
| 11 |
+
- precision
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
## Model Performance
|
| 15 |
+
|
| 16 |
+
The model achieved the following scores on the evaluation dataset:
|
| 17 |
+
|
| 18 |
+
- **Accuracy**: 94.6%
|
| 19 |
+
- **F1 Score**: 94.3%
|
| 20 |
+
- **Recall**: 94.6%
|
| 21 |
+
- **Precision**: 94.5%
|
| 22 |
+
|
| 23 |
+
## Intended Use & Limitations
|
| 24 |
+
|
| 25 |
+
- **Best for:** CCTV footage analysis, anomaly detection
|
| 26 |
+
- **Not suitable for:** Non-surveillance video types, real-time processing with limited hardware
|
| 27 |
+
|
| 28 |
+
## Training Details
|
| 29 |
+
|
| 30 |
+
- **Learning Rate:** 5e-6
|
| 31 |
+
- **Batch Size:** 2
|
| 32 |
+
- **Optimizer:** Adam
|
| 33 |
+
- **Training Steps:** 4176
|
| 34 |
+
|
| 35 |
+
## Framework Versions
|
| 36 |
+
|
| 37 |
+
- Transformers: 4.39.3
|
| 38 |
+
- PyTorch: 2.1.2
|
| 39 |
+
- Datasets: 2.18.0
|
| 40 |
+
|