Tidy the Hack up, as the name suggests we ran into a similar idea of classifying our living rooms as classy or messy.
What it does
It assesses whether the room or space uploaded is messy or clean. It uses the Resnet50 model to check the same. It has an upload button to browse the files and uses the machine learning model to compute a messiness index. The messiness index is how clean the room is. Ironic, huh?
How we built it
We built the frontend with streamlit and the model was made with Google Colab and Fast.ai. We used firebase to store the images and used linode to host the website.
Challenges we ran into
We ran into many challenges. We found it difficult to operate with streamlit and we learned how to use it within a day. We had difficulties in finding the appropriate model and customizing it to meet our needs. We also spent a lot of time collecting the data needed to classify the rooms.
Accomplishments that we're proud of
We are proud of the fact that the web-app is up and running.
What we learned
We learned how to use streamlit, train neural networks, work with Google Colab, prototype with Figma, and also collect and analyze the data.
What's next for Clear the Clutter
Adding more social features to the website and having points and leaderboards for different organizing games.