Hugging Face's logo

jparedesDS
/
cs2-yolo12m-weapon-detection

Object Detection
TensorBoard
Model card Files Metrics Community
jparedesDS commited on
Commit
a3cf402
·
verified ·
1 Parent(s): 3966f9e

Create README.md

Browse files
Files changed (1) hide show
  1. README.md +55 -0
README.md ADDED
@@ -0,0 +1,55 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ # Counter Strike 2 Weapon Detector
2
+
3
+ #### Supported Labels
4
+ ['AK47', 'M4A1-S', 'M4A1', 'GALIL', 'FAMAS', 'TEC-9', 'FIVE-SEVEN', 'GLOCK-18', 'USP-S', 'EAGLE', 'BERETTAS', 'P2000', 'MAC10', 'MP5', 'MP9', 'P90', 'P250', 'SSG08', 'AWP']
5
+
6
+ #### How to use
7
+ ```
8
+ from ultralytics import YOLO
9
+
10
+ # Load a pretrained YOLO model
11
+ model = YOLO(r'weights\cs2-yolo12-weapon-detection.pt')
12
+
13
+ # Run inference on 'image.png' with arguments
14
+ model.predict(
15
+ 'image.png',
16
+ save=True,
17
+ device=0
18
+ )
19
+ ```
20
+
21
+ #### Labels
22
+ ![labels.jpg](https://cdn-uploads.huggingface.co/production/uploads/62e1c9b42e4cab6e39dafc97/TRM3AEKgtpsc4Mq6MwyMx.jpeg)
23
+ #### Results
24
+ ![results.png](https://cdn-uploads.huggingface.co/production/uploads/62e1c9b42e4cab6e39dafc97/toGWnn0SIJ_673BycYOes.png)
25
+ #### Predict
26
+ ![val_batch1_labels.jpg](https://cdn-uploads.huggingface.co/production/uploads/62e1c9b42e4cab6e39dafc97/jP93wc2HietXSBScKUHOG.jpeg)
27
+ ![train_batch2.jpg](https://cdn-uploads.huggingface.co/production/uploads/62e1c9b42e4cab6e39dafc97/NqxdnGl7_1H-Gr6TsOpWG.jpeg)
28
+ ![train_batch0.jpg](https://cdn-uploads.huggingface.co/production/uploads/62e1c9b42e4cab6e39dafc97/S89frPmViNrtLjmEwD61y.jpeg)
29
+ ```
30
+ YOLOv12m summary (fused): 169 layers, 20,119,561 parameters, 0 gradients, 67.2 GFLOPs
31
+ Class Images Instances Box(P R mAP50 mAP50-95): 100%|██████████| 3/3 [00:01<00:00, 2.61it/s]
32
+ all 87 158 0.963 1 0.993 0.871
33
+ AK47 11 14 0.979 1 0.995 0.853
34
+ M4A1-S 8 12 0.987 1 0.995 0.849
35
+ M4A1 9 13 0.978 1 0.995 0.844
36
+ GALIL 8 11 0.974 1 0.995 0.859
37
+ FAMAS 7 7 0.963 1 0.995 0.926
38
+ TEC-9 4 4 0.985 1 0.995 0.917
39
+ FIVE-SEVEN 5 5 0.96 1 0.995 0.903
40
+ GLOCK-18 4 4 0.932 1 0.995 0.905
41
+ USP-S 10 10 0.966 1 0.995 0.9
42
+ EAGLE 4 4 0.933 1 0.995 0.847
43
+ BERETTAS 4 4 0.929 1 0.995 0.948
44
+ MAC10 6 6 0.953 1 0.995 0.891
45
+ MP5 7 11 0.978 1 0.995 0.852
46
+ MP9 7 11 0.978 1 0.995 0.888
47
+ P90 9 13 0.978 1 0.995 0.842
48
+ P250 6 6 0.977 1 0.995 0.861
49
+ SSG08 8 11 0.9 1 0.95 0.785
50
+ AWP 8 12 0.99 1 0.995 0.805
51
+ Speed: 0.2ms preprocess, 9.3ms inference, 0.0ms loss, 1.2ms postprocess per image
52
+ ```
53
+
54
+ #### Others models Counter Strike 2
55
+ https://huggingface.co/ChitoParedes/