Instructions to use zjunlp/Ocean-router with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- ultralytics
How to use zjunlp/Ocean-router with ultralytics:
from ultralytics import YOLOvv11 model = YOLOvv11.from_pretrained("zjunlp/Ocean-router") source = 'http://images.cocodataset.org/val2017/000000039769.jpg' model.predict(source=source, save=True) - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,25 +1,35 @@
|
|
| 1 |
-
---
|
| 2 |
-
language: en
|
| 3 |
-
license: apache-2.0
|
| 4 |
-
library_name: ultralytics
|
| 5 |
-
tags:
|
| 6 |
-
- yolov11
|
| 7 |
-
- yolov5
|
| 8 |
-
- image-classification
|
| 9 |
-
- routing
|
| 10 |
-
- marine-images
|
| 11 |
-
- oceangpt-x
|
| 12 |
-
---
|
| 13 |
-
# Ocean-router: Marine Image Routing Classifiers
|
| 14 |
-
Two-stage lightweight classifiers used to dynamically route marine images to specialized detectors based on image modality and content.
|
| 15 |
-
## Model Files & Tasks
|
| 16 |
-
| File | Task | Architecture | Input/Output |
|
| 17 |
-
|------|------|--------------|--------------|
|
| 18 |
-
| `cls_bio_sonar/best.pt` | Sonar vs. Biological routing | YOLOv11-cls | Image → `[sonar_prob, bio_prob]` |
|
| 19 |
-
| `fish_coral_cls/best.pt` | Fish vs. Coral
|
| 20 |
-
## Usage
|
| 21 |
-
### 1. Sonar/Biological Router
|
| 22 |
-
```python
|
| 23 |
-
from ultralytics import YOLO
|
| 24 |
-
router = YOLO("cls_bio_sonar/best.pt")
|
| 25 |
-
results = router.predict("input.jpg")
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: en
|
| 3 |
+
license: apache-2.0
|
| 4 |
+
library_name: ultralytics
|
| 5 |
+
tags:
|
| 6 |
+
- yolov11
|
| 7 |
+
- yolov5
|
| 8 |
+
- image-classification
|
| 9 |
+
- routing
|
| 10 |
+
- marine-images
|
| 11 |
+
- oceangpt-x
|
| 12 |
+
---
|
| 13 |
+
# Ocean-router: Marine Image Routing Classifiers
|
| 14 |
+
Two-stage lightweight classifiers used to dynamically route marine images to specialized detectors based on image modality and content.
|
| 15 |
+
## Model Files & Tasks
|
| 16 |
+
| File | Task | Architecture | Input/Output |
|
| 17 |
+
|------|------|--------------|--------------|
|
| 18 |
+
| `cls_bio_sonar/best.pt` | Sonar vs. Biological routing | YOLOv11-cls | Image → `[sonar_prob, bio_prob]` |
|
| 19 |
+
| `fish_coral_cls/best.pt` | Fish vs. Coral routing | YOLOv5 | Image → `[fish_prob, coral_prob]` |
|
| 20 |
+
## Usage
|
| 21 |
+
### 1. Sonar/Biological Router
|
| 22 |
+
```python
|
| 23 |
+
from ultralytics import YOLO
|
| 24 |
+
router = YOLO("cls_bio_sonar/best.pt")
|
| 25 |
+
results = router.predict("input.jpg")
|
| 26 |
+
```
|
| 27 |
+
### 2. Fish/Coral Classifier (YOLOv5)
|
| 28 |
+
Requires the official YOLOv5 repo
|
| 29 |
+
```python
|
| 30 |
+
import torch
|
| 31 |
+
model = torch.hub.load("ultralytics/yolov5", "custom", path="fish_coral_cls/best.pt", force_reload=True)
|
| 32 |
+
results = model("input.jpg")
|
| 33 |
+
```
|
| 34 |
+
cls_bio_sonar decides if input is sonar or biological.
|
| 35 |
+
If biological, fish_coral_cls routes to the appropriate species detector.
|