Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
license: mit
|
3 |
+
pipeline_tag: object-detection
|
4 |
+
tags:
|
5 |
+
- yolo
|
6 |
+
- hatsune miku
|
7 |
+
---
|
8 |
+
|
9 |
+
<h1 style="color: #00ffdd; font-size: 2rem;">Dze je miku? 🥵 (where is Miku)</h1>
|
10 |
+
|
11 |
+
Finetuned [YOLOv8s](https://github.com/ultralytics/ultralytics) on 200 labeled Miku photos for 150 epochs, detects a Hatsune Miku doll in an image 🥰.
|
12 |
+
|
13 |
+
Demo WASM deploy (runs in your browser): https://plasmoxy.github.io/dzejemiku
|
14 |
+
|
15 |
+
```py
|
16 |
+
!pip install -U "huggingface_hub[cli]"
|
17 |
+
!pip install ultralytics
|
18 |
+
!huggingface-cli download Plasmoxy/dze-je-miku-yolo miku_yolo.pt --local-dir .
|
19 |
+
|
20 |
+
from ultralytics import YOLO
|
21 |
+
|
22 |
+
model = YOLO('./miku_yolo.pt')
|
23 |
+
|
24 |
+
res = model.predict([
|
25 |
+
'test.jpeg'
|
26 |
+
], save=True, save_txt=True)
|
27 |
+
|
28 |
+
for r in res:
|
29 |
+
for b in r.boxes:
|
30 |
+
print(b.xyxy)
|
31 |
+
|
32 |
+
```
|
33 |
+
|
34 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/6374c3da905caa7e85bbb0c6/AeH2OljyGQ39jZEsixO6W.png" width="500" />
|
35 |
+
<img src="https://cdn-uploads.huggingface.co/production/uploads/6374c3da905caa7e85bbb0c6/YLVJWOlbUM-6u5CfLrDJ2.png" width="500" />
|
36 |
+
|
37 |
+
|