Inspiration
The respected DS3 society.
What it does
When a user inputs an image of a dinosaur, our convolutional neural network identifies the image and returns the name of the dinosaur using our trained database from the Simple Dinosaur dataset. Furthermore, it connects the name of the dinosaur to where it lived and plots the country on a world map.
How we built it
We watched YouTube videos to learn the theoretical concepts and syntax. Additionally, we utilized various online resources such as GeekforGeeks, StackOverflow, Keras official documentation, and Medium.
Challenges we ran into
We faced massive issues in improving the accuracy of the convolutional neural network.
Accomplishments that we're proud of
Making a convolutional neural network for the first time.
What we learned
Learning TensorFlow, Keras and Geopandas.
What's next for DinoMap
Improving its accuracy and incorporating more classes, i.e. more dinosaurs.
Log in or sign up for Devpost to join the conversation.