LSTM_PM10_airpolution_Sofia
LSTM model for prediction of the air polution in Sofia, Bulgaria
This repository contains a python code of an LSTM (Long short-term memory) machine learning model for the prediction of air polution in the city of Sofia, Bulgaria. The model is to be tested soon against a meteo model, created by the National Institute of Meteorology and Hydrology. The model here performs ~30% better than the naive-model "today will be as poluted as yesterday", a standard dummy naive-model in meteorology. The data for training the model are from the official, state operated meteo-stations of IAOS in Sofia.
The code can be also viewed on colab: https://colab.research.google.com/drive/1_UVt6wUKmodF2y7AfFCSo7uWKd6WoH1y?usp=sharing
