michalliu / tf-weather

Weather forecasting with RNN, using TensorFlow

Weather Forecasting

Using TensorFlow and Long Short-Term Memory Recurrent Neural Networks.

Background: research has shown that temperature, humidity, and wind can be accurately predicted out to 72hrs using a 15 year data period of hourly measurements. Mostly based on this paper.

Get started

Set up a Python virtual environment and install requirements via pip:

$ virtualenv venv
$ source venv/bin/activate
$ pip install -r requirements.txt

Package content

Scripts:

To download data form Weather Company Data, use tools/weather_data.py > data.csv. This will download the last year's daily minimum and maximum temperature values for a selected location and save it in data.csv. Before use, make sure to sign up for an account with them and add the username and password to the .env file.

License

Apache v2 License

About

Richard Dancsi

• Blog: wimagguc.com
• Github: github.com/wimagguc
• Twitter: twitter.com/wimagguc
• Linkedin: linkedin.com/in/richarddancsi
• Company: buntlabs.com