Inspiration
As postsecondary students with little experience in our respective fields, it is hard to know how much we should be getting paid, and thus negotiate offers. We decided to build a starting point for a salary visualization tool that in the future will be made much more powerful and help us combat economic inequality.
What it does
We built a linear regression model to predict one's salary around the world given certain parameters. The model looks at stuff such as skills, tools, coding experience... The frontend then takes this information and displays it with data visualization tools.
How we built it
We used the 2019 and 2018 stack overflow annual developers survey as the training data for our model. The dataset was manipulating into a format that made sense for our project. The data manipulation was done primarily with pandas and the regression model was built with scikit-learn. We then used pickle to save and load our regression model into a flask-restful app, allowing us to deploy it wherever we want. Finally, our Frontend was built with React Typescript, the data visualization uses amCharts4.
Challenges we ran into
Our project went pretty smoothly up until we had to deploy it. We ran into a lot of CORS issues which we were stuck on for hours. We eventually ended up running locally. amCharts4 is natively a javascript library, as a result functional components had to be converted to class based components.
Log in or sign up for Devpost to join the conversation.