How to Use
To use this model in your project, follow the steps below:
1. Installation
Ensure you have the ultralytics
library installed, which is used for YOLO models:
pip install ultralytics
# class
calling
clapping
cycling
dancing
drinking
eating
fighting
hugging
laughing
listening_to_music
running
sitting
sleeping
texting
using_laptop
2. Load the Model
You can load the model and perform detection on an image as follows:
from ultralytics import YOLO
# Load the model
model = YOLO("./action-11x.pt")
# Perform detection on an image
results = model("image.png")
# Display or process the results
results.show() # This will display the image with detected objects
3. Model Inference
The results object contains bounding boxes, labels (e.g., numbers or operators), and confidence scores for each detected object.
Access them like this:
# View results
for r in results:
print(r.probs) # print the Probs object containing the detected class probabilities
Inference API (serverless) does not yet support ultralytics models for this pipeline type.