The inspiration for this project came through Pinata even before the hackathon started, as a common interest between all of our team members. After the challenge was given, we knew that this was the challenge we wanted to do.
What it does. MLVerse is a minimalistic ML model management tool that allows users to get in depth analysis and improvements on the ML models that they upload to MLVerse.
How we built it. We built MLVerse using Pinata, as well as using Next js and Clerk for user authentication. Next Js took care of our routing systems, and clerk took care of our user sign in and sign out functions.
Challenges we ran into. We ran into routing challenges with the upload feature, and we had to reqork the routing system that we initially coded, finally fixing it and getting it to work.
Accomplishments that we're proud of. We are proud that we were able to create such a minimalistic and asthetic ML model management system that provides the user with easy access to the errors and improvements that they can make on their models. Compared to other tools, MLVerse is the first of its kind and a step towards the future.
What we learned. We learned that using Pinata is a very hard tool to use, but also very very useful and makes our coding time a lot easier. Running into errors doesn't mean you should give up, it means you should keep trying, ask for help if you need it and fix them.
What's next for MLVerse. We are planning on adding a compare and contrast function for MLVerse which will allow users to comapre and contrast the different modles that they upload
Log in or sign up for Devpost to join the conversation.