Spaces:
Sleeping
Sleeping
ux updates
Browse files
app.py
CHANGED
@@ -8,7 +8,7 @@ yolov5_dff = os.path.join(os.getcwd(), "data/xai/yolov5_dff.png")
|
|
8 |
yolov8_dff = os.path.join(os.getcwd(), "data/xai/yolov8_dff.png")
|
9 |
|
10 |
architecture_description_yolov5 = """
|
11 |
-
|
12 |
|
13 |
- **Backbone**: Uses **CSPDarknet53** for feature extraction with **ResNet**-like residual connections.
|
14 |
- **Neck**: **PANet** and **FPN** aggregate features at multiple scales.
|
@@ -23,7 +23,7 @@ architecture_description_yolov5 = """
|
|
23 |
"""
|
24 |
|
25 |
architecture_description_yolov8s = """
|
26 |
-
|
27 |
|
28 |
- **Backbone**: Uses **CSPDarknet** with efficient feature extraction layers.
|
29 |
- **Neck**: Incorporates **FPN** and **PANet** for multi-scale feature aggregation.
|
@@ -35,6 +35,7 @@ architecture_description_yolov8s = """
|
|
35 |
- **Multi-Task**: Supports tasks like classification, detection, and segmentation.
|
36 |
- **High Accuracy**: Achieves state-of-the-art results on small and medium datasets.
|
37 |
- **Lightweight**: Efficient architecture for smaller models with reduced computational cost.
|
|
|
38 |
"""
|
39 |
|
40 |
|
|
|
8 |
yolov8_dff = os.path.join(os.getcwd(), "data/xai/yolov8_dff.png")
|
9 |
|
10 |
architecture_description_yolov5 = """
|
11 |
+
### YOLOv5 Architecture
|
12 |
|
13 |
- **Backbone**: Uses **CSPDarknet53** for feature extraction with **ResNet**-like residual connections.
|
14 |
- **Neck**: **PANet** and **FPN** aggregate features at multiple scales.
|
|
|
23 |
"""
|
24 |
|
25 |
architecture_description_yolov8s = """
|
26 |
+
### YOLOv8s Architecture
|
27 |
|
28 |
- **Backbone**: Uses **CSPDarknet** with efficient feature extraction layers.
|
29 |
- **Neck**: Incorporates **FPN** and **PANet** for multi-scale feature aggregation.
|
|
|
35 |
- **Multi-Task**: Supports tasks like classification, detection, and segmentation.
|
36 |
- **High Accuracy**: Achieves state-of-the-art results on small and medium datasets.
|
37 |
- **Lightweight**: Efficient architecture for smaller models with reduced computational cost.
|
38 |
+
|
39 |
"""
|
40 |
|
41 |
|