Inspiration
we wanted to help any players of fantasy hockey to make better decisions. We wanted to turn raw data into simple and proper instructions anyone could understand.
What it does
the app turns data into informations that user can ask about to guide them into making the best decisions possible. the app can also manage a team, put players on a bench, deals with injured players
How we built it
the app is powered by python for data processing, we used pandas for normalization and cleaning of data, steamlit has been used for an easy to use and fast interface, Gemini was used for the chatbot, machine learning to develop a solid model that gives the most precise prediction possible and we used yahoo FantasyAPI as the main data source.
Challenges we ran into
machine learning and connection our two API were by far the most difficult task we encountered but after a lot of effort , we accomplished it. Data normalization was also challenging since it was pretty messy.
Accomplishments that we're proud of
We develop an algorithm who's pretty precise after training it. We also managed to connect both Gemini and fantasyAPI so it can respond more precisely
What we learned
Machine learning is a tedious task that takes a lot of time to be effective, Giving Prompts to gemini is also harder than we imagined
What's next for Hockey-helper
add an algoritm that can also predict goalkeeper and to be able to deploy our project
Built With
- gemini
- machine-learning
- pandas
- python
- yahoofantasyapi
Log in or sign up for Devpost to join the conversation.