Inspiration
Our inspiration was how heavily sleep deprivation affected our daily lives, academically, physically, and mentally. We constantly had a lack of energy and focus and this led to a steep decline in performance. This project serves to determine sleep quality before you sleep, so you can quickly determine your performance for the following day and plan accordingly.
What it does
Our project takes inputs from the user such as hours of intended sleep, age, gender, caffeine consumption, alcohol consumption, and smoking status, and uses them to determine the predicted percentages of light, REM, and deep sleep. These predictions are then used to determine sleep quality; typically a good night of sleep consists of at least 7 hours with 50-60% light sleep, ~25% REM sleep, and ~25% deep sleep.
How we built it
We constructed a simple and easy-to-use website that describes what a good nights sleep consists of, has a tab where the ML model is implemented for the user, and an about us tab.
Challenges we ran into
We ran into numerous challenges, the most notable being actually implementing the ML model into the website, constructing the ML model to produce multiple variable predictions as opposed to just one, and formatting the website to make it accessible for all users.
Accomplishments that we're proud of
Accomplishments we are proud of are actually finishing a furnished website with complete functionality, creating the ML model, which was adequately complex, and creating a easy-to-use website for all users.
What we learned
We learned a lot of technical info about ML and HTML, but also a lot about patience and perseverance. There were many roadblocks along the way, but with time it became evident that it was quintessential to keep a level head.
What's next for Soothing Sleepers
The next steps for Soothing Sleepers is to take into account previous nights of sleep, and form a general sleep pattern calendar as well as a recommended and feasible sleep pattern calendar for the user to follow to get their sleep schedule back on track as efficiently as possible.
Log in or sign up for Devpost to join the conversation.