Trash Classifier

Inspiration

The inspiration for this came when we learned that in Japan they split up all the recyclables so nobody has to later on. This reduces the cost by a lot and leads to higher recycling overall. In the US, people just clump stuff together mostly because they deem splitting up too much work. So we came up with a way for the computer to do this using modern ML technology.

What it does

Given an image, it classifies it into cardboard, glass, metal, paper, or trash.

How we built it

We used the colab feature of google to make the model. First, the data was cleaned and split up into a train, test, and validate set. Next, a CNN model was attempted to be created to classify.

Challenges we ran into

We did not know how a CNN model worked so unfortunately we got stuck and could not finish that part. The model was not fully created due to this.

Accomplishments that we're proud of

We are proud of how far we got and of the friendships we formed.

What we learned

We learned how to create an ML model from scratch for any situation. We also learned the importance of choosing the correct model, otherwise, it could take too long.