This project aims to develop an AI engine that teaches people how to sort trash with computer vision techniques.

I used a pre-trained residual convolutional neural network (resnet50) and, through transfer learning, trained the last layer with a dataset of 2,500 images of trash, obtaining an accuracy of 91% in the test set. Then, I implemented a hardware prototype to turn on LED lights according to the predicted category of trash.

Created at: May 14, 2020              Last modified: May 14, 2020

---

In this project, I tried to predict rental prices of Airbnb listings from features related to the properties, such as amenities, rooms and location.

I trained different regression models (Linear Regression, Decision Tree and Random Forest) using the python scikit-learn library, and then I comapred the performance of them through cross-validation.

Created at: May 19, 2020              Last modified: June 4, 2020

---