README

A general obect detection class based on yolo darknet model.

Inputs

• model(str): yolo model name (default:'yolo')
• threshold(float): threshold of the given model (default:0.4)
• gpu(float): percentage of gpu use desired (default:0 running entirely on cpu)

Requirements

• tensorflow
• cython
• darkflow
• git-lfs

Installation

pip install --user cython
pip install --user git+https://github.com/MacherLabs/darkflow.git

pip install --user git+<https-url>

or

pip install --user git+ssh://[email protected]/macherlabs/yolo.git

Usage

from yolo import ObjectDetectorYolo
detector = ObjectDetectorYolo(model='yolo')
image_url = '../test.jpg'
imgcv = cv2.imread(image_url)
if imgcv is not None:
    results = detector.run(imgcv)
    pprint.pprint(results)
else:
    print("Could not read image: {}".format(image_url))# yolo