purr. is the social networks made by cats
cd purr.frontend
npm install
cd purr.backend
composer install
cp .env.example .env
# Config the environment variables
php artisan key:generate
php artisan migrate
#php artisan db:seed
The image engine is implemented as a Flask API and uses Gunicorn as the WSGI HTTP server for production environments.
Required Python Version: The application requires Python version 3.7 or higher.
Tested Python Versions: The following Python versions have been tested and are confirmed to be compatible:
- 3.11.9
Warning
Python version 3.12 is currently not supported by this configuration of Gunicorn.
cd purr.imageEngine
python -m venv .venv
. .venv\bin\activate # .\.venv\Scripts\activate on Windows
pip install -r requirements.txt
purr. uses a custom model which is pretrained on the YOLOv8 architecture. This model is designed to provide high accuracy for object detection tasks.
Build the Docker image using the provided Dockerfile located at purr.imageEngine/ai/Dockerfile. This image includes all necessary dependencies, including GPU support for training models.
docker build -t pytorch_jupyter -f ./purr.imageEngine/ai/Dockerfile.jupyter .
To run the Docker container with GPU support and access to Jupyter, use the following command:
docker run --gpus all -p 8888:8888 -v purr.imageEngine/ai:/workspace -it pytorch_jupyter
You may use an absolute path for mapping the directory on Windows.
After running the container, Jupyter server should be accessible via http://localhost:8888.
To begin training the model, navigate to the Jupyter Notebook provided:
- Open your browser and go to http://localhost:8888 to access Jupyter.
- Navigate to the notebooks directory.
- Open the train.ipynb notebook.
- Follow the instructions within the notebook to start the training process.
purr. never gonna give you up, never gonna let you down, never gonna run around and desert you. purr. never gonna make you cry, never gonna say goodbye, never gonna tell a lie and hurt you.
- Rick Catsley (2024)