Inspiration
I took inspiration from Hugging face's auto train.
What it does
It generates ML models completely online with out coding or manual training setup.
How we built it
We used next js as front end frame-work while used python for backend, besides that pytorch for model architecture, supabase for DB, Hugging face for deploying models, local/cloud server for training.
Challenges we ran into
The biggest challenge was to make a proper hugging face deployment architecture.
Accomplishments that we're proud of
We've successfully created a fully autonomous no code machine learning platform.
What we learned
We've learned the core knowledge of data mining.
What's next for Pipeline | World's first no code machine learning platform
We're soon going to launch it as a official product for early users.
Built With
- next-js
- pytorch
- supabase
Log in or sign up for Devpost to join the conversation.