Unique Guess

• 

  Cool 30 day prediction on NVDA(on orange)

• 

  A train and test on NVDA stock, blue for the set we train, orange for actual rate or a test set and greed for predict, pretty close

Inspiration

The inspiration came from one of the sponsor which is Fidelity investment. At that time we have no idea what topic to do and the investment came into our mind we our most familiar topic is data analyzing so we plan to make something crazy, which is predicting stocks.

What it does

We use data from yfinance python library where i can get the stock closed price from years ago until now. Then we research for a good AI/ML algorithm that do some good job in predicting stocks. And we choose Long short term memory as our model to train the stock data. We have created two different python files one split the data into train set and test set to see how accurate is the prediction. Another train data from 2020 until now and predict for future.

How we built it

We build it using existed python libraries and with help with copilot!!(Thank UChicago to have a copilot workshop, it helps a lot in debugging and save our time to search in stackoverflow) Also when we stuck or don't know how to continue, it can write code for us by just writing or explaining what we want. By the way this is our first time using LSTM, before that we don't even know it exist!!

Challenges we ran into

We came in really late to the hackathon and at first we have no idea on what to do. Our team have tried making game with a fantastic background music from pyxel library(workshop on Friday) and also tried doing front end using react language and aut0 as a login page. But both of the project, we stuck at coding because our team have lack of experience in front end and the part or API, Domain and web socket linking stuff. We want to make use of some sponsor's API and make a website but we have no experience with it, even react we learn on Friday's workshop. Lastly we just came up with something we are most familiar with which is python in data analysis.

Accomplishments that we're proud of

We are proud of finishing this project!!!

What we learned

we have learned how to use pyxel,react and copilot from the workshops. We learn how to use tensorflow libraries and how to implement LSTM.

What's next for Unique Guess

Be Unique..........

Built With

• python