AQI predictions

model v1.0.0 // Accuracy: 76.34% // last trained: 2024-05-12

Predicting AQI values in Montreal, QC, over the last year, using a deep learning model and achieving an overall accuracy of 76.34%. The model was able to extrapolate patterns in AQI values from historical hourly data. In addition to past air quality index, temperatures, humidity and precipitations were included to improve the model's accuracy.

The data was sourced from the city of Montreal (french) and the federal climate data. The model was trained on a dataset spanning from 2014 to 2022, then the rest was used to validate and assess accuracy of predictions.

Check out the Jupyter notebook. I've also written a blog post about how I made this project.