Update README.md
Browse files
README.md
CHANGED
|
@@ -74,11 +74,11 @@ Each model outputs **9 classes** (species IDs in `metadata/classes.csv` within t
|
|
| 74 |
### Class weights variant
|
| 75 |
Inverse-frequency class weights slightly improved **Spinytail Skate** but reduced aggregate mAP.
|
| 76 |
|
| 77 |
-
|
| 78 |
|
| 79 |
> These weights are YOLOv7-format PyTorch checkpoints. Use the YOLOv7 repository or a compatible runner.
|
| 80 |
|
| 81 |
-
|
| 82 |
|
| 83 |
```bash
|
| 84 |
# 1) Clone YOLOv7 (example URL; use the official repo you trained with)
|
|
@@ -93,4 +93,17 @@ python detect.py \
|
|
| 93 |
--img-size 640 \
|
| 94 |
--conf-thres 0.25 \
|
| 95 |
--iou-thres 0.65 \
|
| 96 |
-
--save-txt --save-conf
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 74 |
### Class weights variant
|
| 75 |
Inverse-frequency class weights slightly improved **Spinytail Skate** but reduced aggregate mAP.
|
| 76 |
|
| 77 |
+
### Quick inference
|
| 78 |
|
| 79 |
> These weights are YOLOv7-format PyTorch checkpoints. Use the YOLOv7 repository or a compatible runner.
|
| 80 |
|
| 81 |
+
## CLI (YOLOv7)
|
| 82 |
|
| 83 |
```bash
|
| 84 |
# 1) Clone YOLOv7 (example URL; use the official repo you trained with)
|
|
|
|
| 93 |
--img-size 640 \
|
| 94 |
--conf-thres 0.25 \
|
| 95 |
--iou-thres 0.65 \
|
| 96 |
+
--save-txt --save-conf
|
| 97 |
+
```
|
| 98 |
+
### Citation
|
| 99 |
+
|
| 100 |
+
If you use these models or the dataset, please cite both:
|
| 101 |
+
|
| 102 |
+
Dataset citation (plain text):
|
| 103 |
+
Morris, C. J., Ayyagari, K. D., Porter, D., Nguyen, Q. K., Hanlon, J., & Whidden, C. (2025).
|
| 104 |
+
Newfoundland Marine Refuge Fish Classification Dataset (N-Marine).
|
| 105 |
+
Government of Canada Open Data Portal.
|
| 106 |
+
https://open.canada.ca/data/en/dataset/2ae46860-f82a-4127-bb1f-b02e36ef6a70
|
| 107 |
+
|
| 108 |
+
Model citation (plain text):
|
| 109 |
+
TODO: Add
|