Predicting extreme weather temperatures

Welcome to the forefront of extreme weather temperature prediction, where artificial Intelligence (AI) takes center stage, delivering unparalleled accuracy and insights. Extreme weather events, such as heatwaves, cold spells, and storms, have far-reaching consequences on human lives, infrastructure and ecosystems. By harnessing the potential of AI, we can now unlock the ability to forecast with an unprecedented level of precision and advance notice to prepare, adapt and mitigate risks associated with several weather events or to optimize business operations.


  • This use case is based on the “Extreme-Weather Temperature Prediction” dataset from Kaggle.
  • The goal was to predict the next day’s maximum air temperature using 21 values including the current day’s min and max temperature, latitude, longitude, elevation, slope, and solar radiation.
  • We used NanoEdge AI Studio to create an Extrapolation project based on these inputs that could achieve this goal.


Temperature sensors and generic sensors.


Extrapolation targets: Next day maximum air temperature
Signal length: 21 (multi sensors)
Data rate: every day


90.18% accuracy, 0.4 Kbytes of RAM, 3.6 Kbytes of Flash memory