Climatewatch Insights

A Time-series analytics dashboard built to forecast and visualise data associated with socio-economic indicators and historic emission of 140 nations. RNNs have been used to perform forecasting. Seasonal decomposition and comparison of autocorrelation methods along with an API version has been implemented to detect anomalies. Kalman filters have been used to estimate the best developing regions by combining various emission indicators.

About Climatewatch

ClimateWatch is an open online platform designed to empower users with the climate data, visualizations and resources they need to gather insights on national and global progress on climate change, sustainable development, and help advance the goals of the paris agreement.

Run Locally

Clone the project

  git clone https://github.com/DeepanNarayanaMoorthy/ClimateWatch-Insights.git

Go to the project directory

  cd ClimateWatch-Insights

Install dependencies

  pip install -r requirements.txt

Start the server

  streamlit rum dashboard.py

Acknowledgements

• ClimateWatch Data Source

Tech Stack

Client: StreamLit

Server: Python, scikit-learn, Tensorflow