Skip to content

Precipitation forecast using machine learning time series algorithms

Notifications You must be signed in to change notification settings

ManuelS0usa/TS-Precipitation-Forecast

Repository files navigation

Precipitation Forecast using Machine Learning Time Series Algorithms

Time Series Machine Learning Project | Manuel Sousa | 25.04.2022

Context

Viveiros Monterosa [http://www.monterosa.pt/pt/] are a flower production company from the south of Portugal, more specifically, Moncarapacho, Olhão. To produce their plants the company have their production fields equipped with multiple sensors in order to efficiently control and understand the yearly climate enviroment. This project takes advantages of their precipitation sensors collected data, grouped by month.

Scripts

The following scripts were built in the scope of the development of this project:

  • data_load.py: Class to handle data importing and transformation;
  • eda.ipynb: Exploratory data analysis;
  • arma_models.ipynb: Time series arma and holt-winters based models;
  • rnn_models.ipynb: Recurrent neural networks (LSTM) time series based models;
  • prophet_model.ipynb: Facebook Prophet based model.

About

Precipitation forecast using machine learning time series algorithms

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published