Update README.md
Browse files
README.md
CHANGED
|
@@ -29,7 +29,7 @@ This repository provides a **compressed version of YOLOv5s** using [AIminify](ht
|
|
| 29 |
The base architecture is **YOLOv5s** from Ultralytics. Modifications include:
|
| 30 |
- **Pruning** certain channels/kernels based on AIminify’s pruning algorithm.
|
| 31 |
- **Automatic Fine-Tuning** post-pruning to recover performance.
|
| 32 |
-
|
| 33 |
Despite its reduced size, this model maintains similar detection capabilities for common objects as the original YOLOv5s.
|
| 34 |
|
| 35 |
## Intended use
|
|
@@ -37,7 +37,7 @@ Despite its reduced size, this model maintains similar detection capabilities fo
|
|
| 37 |
- **Primary use case**: General object detection (people, vehicles, animals, etc.) in images and videos.
|
| 38 |
- **Industries**: Could be applied in retail (store analytics), security, robotics, autonomous vehicles, or any scenario where a fast, lightweight detector is beneficial.
|
| 39 |
- **Resource-constrained environments**: Ideal for devices or deployments where GPU/CPU resources are limited or when high throughput is required.
|
| 40 |
-
|
| 41 |
### Limitations
|
| 42 |
|
| 43 |
- **Dataset bias**: Trained on COCO, which may not generalize to highly domain-specific use cases (e.g., medical imaging, satellite imagery). Additional domain-specific fine-tuning might be necessary.
|
|
|
|
| 29 |
The base architecture is **YOLOv5s** from Ultralytics. Modifications include:
|
| 30 |
- **Pruning** certain channels/kernels based on AIminify’s pruning algorithm.
|
| 31 |
- **Automatic Fine-Tuning** post-pruning to recover performance.
|
| 32 |
+
|
| 33 |
Despite its reduced size, this model maintains similar detection capabilities for common objects as the original YOLOv5s.
|
| 34 |
|
| 35 |
## Intended use
|
|
|
|
| 37 |
- **Primary use case**: General object detection (people, vehicles, animals, etc.) in images and videos.
|
| 38 |
- **Industries**: Could be applied in retail (store analytics), security, robotics, autonomous vehicles, or any scenario where a fast, lightweight detector is beneficial.
|
| 39 |
- **Resource-constrained environments**: Ideal for devices or deployments where GPU/CPU resources are limited or when high throughput is required.
|
| 40 |
+
|
| 41 |
### Limitations
|
| 42 |
|
| 43 |
- **Dataset bias**: Trained on COCO, which may not generalize to highly domain-specific use cases (e.g., medical imaging, satellite imagery). Additional domain-specific fine-tuning might be necessary.
|