tae898/yolov5 repository overview

⁠YOLO v5 RESTful API server

This repo is forked form https://github.com/ultralytics/yolov5⁠. It's nothing but a RESTful API server that is dockerized.

⁠app.py⁠

This is the flask api server. It returns the following values:

results = [
    {
        "bbox": [int(r) for r in result[:4]],
        "det_score": result[4],
        "label_num": int(result[5]),
        "label_string": names[int(result[-1])],
    }
    for result in results
]

⁠docker

Checkout the Dockerfile⁠.

You don't have to build the docker image. You can just pull it from the docker hub⁠.

docker pull tae898/yolov5

This repo contains the weight file yolov5s.pt so that docker does not need an Internet access to download the weights.

⁠How to make a client

You have to write a client in python. Below is an example.

import jsonpickle

image_path = './image.jpg'

with open(image_path, "rb") as stream:
    binary_image = stream.read()

data = {"image": binary_image}

to_send = jsonpickle.encode(data)
response = requests.post(url_yolo, json=to_send)
response = jsonpickle.decode(response.text)
results = response["yolo_results"]

I know that this is not the most standard way to do such a thing but this was the eaiset way for me somehow. jsonpickle works really well with different types of python objects (e.g., dict, numpy, binary string, etc.)

⁠Troubleshooting

The best way to find and solve your problems is to see in the github issue tab. If you can't find what you want, feel free to raise an issue. We are pretty responsive.

⁠Cite our work

⁠Contributing

Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.

1. Fork the Project
2. Create your Feature Branch (git checkout -b feature/AmazingFeature)
3. Commit your Changes (git commit -m 'Add some AmazingFeature')
4. Push to the Branch (git push origin feature/AmazingFeature)
5. Open a Pull Request

⁠Authors

• Taewoon Kim⁠