This project utilizes machine learning to predict extreme weather events and analyze traffic congestion in Southeast Asia. The repository includes data analysis, model training, and visualization scripts to forecast weather conditions for the next 30 days and enhance transportation efficiency.
GNU General Public License v3.0
11
stars
1
forks
source link
Create a forecasting model to predict temperature on open-meteo's weather data #2 #36
Open-meteo data is located in the folder data/weather in a file called open_meteo.csv. Some of the attributes have information on temperature. You can check the meaning of all the attributes below:
What you'll do:
Select a city from the ones available, extract the time series of temperature values from this city
Separate the last three years of data as test data and use the rest to train a forecasting model
Compare your forecasting model's prediction to the test data
Feel free to use any technique you wish to work with. An easy to start is Meta's Prophet, but that's really up to you, you can use something popular like ARIMA or whatever.
Open-meteo data is located in the folder
data/weatherin a file calledopen_meteo.csv. Some of the attributes have information on temperature. You can check the meaning of all the attributes below:What you'll do:
Feel free to use any technique you wish to work with. An easy to start is Meta's Prophet, but that's really up to you, you can use something popular like ARIMA or whatever.
Place your work in the
src/weather/modelsfolder.