Inspiration
Having participated in sports trials at university, we decided to implement a solution based on our common interest in sports.
What it does
Simplifies the role of sports scouts by taking statistics in and providing a playstyle as to what the prospects of the players are through the use of AI and ML.
How we built it
Used a tech stack of the IBM Cloud VM, Docker and Jupyter Labs, as well as GitHub to store source code and datasets.
Challenges we ran into
We found that the playstyles were subjective, and the initial dataset we used was not up to the best standard for our application, so we transitioned into predicting goals scored, a metric that could be measured against quantitative stats for accuracy.
Accomplishments that we're proud of
We learnt to use a VM, Docker and Jupyter Notebook tech stack, as well as working in part of a team with Jupyter Collaboration plugin. We also managed to implement a ML model that achieved 74% in very little time, especially with changes to our project’s aims partway through.
What we learned
A lot of data science is done prior to the design of the machine learning model, in the data collection and cleaning phases, which we hadn’t had the most experience in before. As well as this, we found that the tech stack plays a vital role in the software development industry, as it allows for testing in a controlled environment.
What's next for The three electrons - 18
We hope to develop our skill as software engineers, data scientists and our ability to design our own software solutions, something not present in our academic courses.
Log in or sign up for Devpost to join the conversation.