-
-
This image describes the machine learning model for galaxy classification that we could not finish.
-
-
This is our actual model which is a CNN model.
-
Inspiration
Our inspiration came from our love of NASA and childhood dreams to explore space and unravel the mysteries beyond our reach.
What it does
Our website allows the user to explore the outer space through interactive modules. They can classify galaxies, different types of stars, and predict if there are exoplanets in a certain image.
How we built it
We built the website in Replit using HTML and CSS and the three machine learning models in google colab which we then transferred two of them to a member's github account. The machine learning models were built using Python, Tensorflow and Scikit-learn.
Challenges we ran into
We could not finish a model regarding galaxy classification as the data we planned on using was too large to be put into Github so we included the Google Colab link to the model and what we have so far.
Accomplishments that we're proud of
We are proud that we could work together on such a fulfilling project and learn new skills. Some members learned more about machine learning while others worked on website-development and learned more about that.
What we learned
We learned what we could do when we work together and how we can contribute to a meaningful project. We also learned Machine Learning models such as CNN and Support Vector Machine models as well as various tools for website development.
What's next for EX0PLAN33R
Next, we will transfer the website to a independently hosted website and continue developing it even further. We will also potentially change it from HTML, CSS to Javascript React. In addition, we will complete the models and create a backend so that the user could use the model on the website itself, rather then having to go to the github and access the models there.
https://drive.google.com/file/d/1_zymzYHRtVUyFAvEopTosW33wz2l5viS/view?usp=sharing
Log in or sign up for Devpost to join the conversation.