metadata
license: mit
base_model:
- Ultralytics/YOLOv8
pipeline_tag: object-detection
tags:
- cardilla
- weed
- yolo
Model characteristics
- Tiled-based inference mode. Optimal input must be 640x640px. Batch-based inference is recommended.
Model Download and Inference
Install dependencies:
pip install ultralytics huggingface_hub
And then run the following code for a simple image inference. You can do the inference with either images or videos, following the Ultralytics convention.
from huggingface_hub import hf_hub_download
from ultralytics import YOLO
model_path = hf_hub_download(
repo_id="jmartinpizarro/carlav8t",
filename="carlav8t.pt"
)
model = YOLO(model_path)
# Perform object detection on an image
results = model("test.png") # Predict on an image, assume 640x640 resolution
results[0].show() # Display results