As a team of mixed experience we had different inspirations for this project. One member wanted to utilise his machine learning skills in a real life situation, one wanted to learn more about python and machine learning, and one wanted to learn more about machine learning while working on a project they would be interested getting a job in.

What it does

The project takes in test data containing past volume and sales and tests it against if the price of the stock went up or down the next day. The data must first be cleaned and the relevant data selected. The model then learns from this data and is tested against an unseen set.

How we built it

We coded in python using PyTorch. The team tried different model implementation and worked together to communicate what patterns they found in order to build the best model.

Challenges we ran into

Dealing with missing data, deciding what is relevant to the stock value, getting the model the generalise across all the data.

Accomplishments that we're proud of

What we learned

We have learnt about integrating financial logic into the machine learning model

What's next for QRT Stock Prediction

Built With

Share this project: