Although bees are tiny, they have a very significant role in the Ecosystem – they are a keystone species. Bees are responsible for pollinating about a third of the agricultural crops. So, you can imagine what would happen if the population of bees drops drastically – many species could go extinct [2,3]. Beekeeping is a good way to maintain and even increase the bee population. Regularly checking up on the bees and hives could help in letting the bee population thrive [1]. The objective of this project is to classify a bee as healthy or not based on its image. Each datapoint in the dataset is a combination of the pixel representation of the bee’s image, and its classification label – 0 or 1 (unhealthy or healthy bee, respectively). This makes it a supervised binary classification problem. The dataset, called “The BeeImage Data: Annotated Honey Bee Images”, was created by Jenny Yang and the 2nd version of the dataset was posted to Kaggle in 2018. The dataset was created by extracting images of bees from time-lapse videos (submitted by other people) of bees around their respective beehives. [1]