Inspiration
As we all know, climate change is one of the biggest problems on our planet. We wanted to create something that would allow anyone to be able to simulate what a statewide temperature change would be like, as a result of a change in the amount of carbon dioxide emissions in their state. Even very small and slight temperature changes can cause drastic efforts on the environment such as floods, droughts, ocean acidity, and wilder weather patterns. Regi can predict these changes and allow people to find an emission rate that will be much safer and reduce the amount of temperature changes.
What it does
The program uses a linear regression model to predict a change in the statewide temperature, based on the previous year's temperature, and change in carbon dioxide emission rate, from the previous, to the current year. The data for this is contained in separate spreadsheets that the user can edit to create simulations on what a particular carbon rate will do to the temperature.
How we built it
We used the scikit learn machine learning library to create our model. We also used the pandas library for data manipulation.
Challenges we ran into
It took us a large majority of our time just to find the necessary data. On top of this, we also had to iterate our project idea multiple times due to failures in algorithms, data manipulation, and more.
Accomplishments that we're proud of
We are proud of the fact that even with our failures, we were able to overcome our struggles and create a viable and innovative solution to our problem. We are also proud of our teamwork in this project and how we overcame our disagreements to create the product we have today.
What we learned
We learned a lot about machine learning and data manipulation. We learned how to create various models to predict trends and how to overcome struggles we faced.
What's next for Project Regi
Our next steps in this project would be to make our program more user friendly and stop users from having to input data in an external file and instead allow them to input data directly. We would also like to deploy Ragi on a server and allow anyone to use it from a web browser or a mobile device.
