deadlock-yolo11l / README.md
jparedesDS's picture
Update README.md
5bfbeae verified
metadata
base_model:
  - jparedesDS/deadlock-yolo11l
tags:
  - yolo11
  - deadlock
  - videogames

Deadlock Players Detector

Supported Labels

['deny', 'enemy', 'hit', 'minions']

ALL my models YOLO11, YOLOv10 & YOLOv9

How to use

from ultralytics import YOLO

# Load a pretrained YOLO model
model = YOLO(r'weights\yolo11l_deadlock.pt')

# Run inference on 'image.png' with arguments
model.predict(
    'image.png',
    save=True,
    device=0
    )

Confusion matrix normalized

confusion_matrix_normalized.png

Labels

labels.jpg

Results

results.png

Predict

val_batch1_labels.jpg

YOLO11l summary (fused): 464 layers, 25,282,396 parameters, 0 gradients, 86.6 GFLOPs
                 Class     Images  Instances      Box(P          R      mAP50  mAP50-95): 100%|β–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆβ–ˆ| 8/8 [00:03<00:00,  2.37it/s]
                   all        551        661      0.874      0.798      0.856      0.486
                  deny         13         14      0.748      0.857      0.819       0.48
                 enemy        197        228      0.942      0.785      0.883      0.508
                   hit         40         53      0.861      0.704      0.787      0.409
               minions        222        366      0.945      0.847      0.933      0.546

Others models Counter Strike 2 YOLOv10m Object Detection

https://huggingface.co/jparedesDS/valorant-yolov10b