research-william created by GitHub Classroom

Current weather forecasts rely mainly on physical simulation and our knowledge of meteorological events (Science Daily). After collecting a variety data from observation, both historical and present, meteorologists feed the data into a model. This model is created from past data using statistical analysis. From this model, a forecast is produced. Another approach that has not yet been fully explored is that of using a convolutional neural network. The goal of this project is not to create a new model, but to show that weather forecasting can also be accomplished with a convolutional neural network.

The programs are written in Python 3.

Much of the project runs on PyTorch, a Python package. To begin, pip install PyTorch.

pip install torch

End with an example of getting some data out of the system or using it for a little demo

• William Wang - Initial work - incognitofox