Restless Learning

Inspiration

We wanted to build something that we wish we had when we were learning.

What it does

It lets people play with a neural network by being able to change how many layers it has and how much nodes per layer. It builds the neural network and lets u train it one epoch at a time so you can see it learning in real time as the accuracy goes up. It uses the MNIST dataset to make its predictions on. It also gives the user tips powered by GPT and teaches them about neural networks as they progress through the epochs during training.

How we built it

We built it using a combination of technologies. We used python with fastapi as the back end and react as the frontend.

Challenges we ran into

Getting the backend and frontend to communicate reliably was one hurdle we overcame. We also learned that the website had to be optimized for various screensizes as some elements would overlap on small screen sizes. We fixed all these issues and overcame the challenges.

Accomplishments that we're proud of

We are very proud that we have a working app. We had a good vision and worked very hard to bring it to reality within the time constraint.

What we learned

Webdev is very deep rabbit hole. But it also unlocks a whole new horizon where users can use ur app with the click if a button without and installations.

What's next for Restless Learning

One goal would be to add more datasets.

Built With

• fastapi
• javascript
• python
• react