Groundhog

Inspiration

Everyone wants to be able to predict the future, so we decided to do some research into how we could use machine learning to try and predict future prices for stocks. Eventually, we discovered LSTM (Long short-term memory), and decided to try and use that model for our task. We spent a lot of time trying to figure out and understand how LSTMs work, and we were able to get a pretty good understanding.

What it does

Given the recent price history of a stock for several days (including opening, high, low, closing, and trading volume), it attempts to predict the price of the stock several days into the future

How we built it

We used the build in LSTM model as well as other tools in keras

Challenges we ran into

The biggest challenge was getting accurate results. Throughout development, our outputs were all over the place but we were eventually able to get something reasomable

Accomplishments that we're proud of

What we're most proud of is the knowledge we gained and a deeper understanding of what LSTMs are, how they work, and when we can use them in the future.