This is the code and data associated with the paper titled:

"Enhancing PM2.5 Prediction Using NARX-Based Combined CNN and LSTM Hybrid Model"

The paper is published in Sensors Journal available at https://www.mdpi.com/1424-8220/22/12/4418

DOI: 10.3390/s22124418

Please cite the paper if you use the code.

If you have any questions, please post them in the Discussions Section.

The code requires Python 3.11. TensorFlow currently does not provide wheels for Python 3.12, so please use Python 3.11 when installing the dependencies.

Download and install Graphviz from:
👉 https://graphviz.org/download/
Make sure to add the Graphviz bin directory to your PATH environment variable.
Install it before running the code.

If you are using a Debian-based Linux environment, you can

install Graphviz with apt:

sudo apt-get update && sudo apt-get install -y graphviz

Requirements

The libraries used in the code are:
tensorflow==2.18.1
keras==3.10.0
pandas==2.2.3
fireTS==0.0.9
xgboost==3.0.2
python-box==7.3.2
python-aqi
pydot
graphviz
matplotlib==3.10.3
numpy==1.26.4
scikit-learn==1.2.1
setuptools>=80
wheel

Install the dependencies with:

pip install -r requirements.txt

If you are using CodeSpaces, the devcontainer will automatically install the dependencies for you.

Edit parameters in airquality/config.py then run python main.py.