Inspiration
We wanted to combine our interest in neural networks, backend development with sence of humor. As well as kickstart our personal project.
What it does
Categorizes images based on their text content. Allowing easy searching in image based content.
How we built it
We built it using Google Cloud Vision and Tensorflow libraries, hacked together via Flask and Vue.js.
Challenges we ran into
We had to create our own dataset of meme classifications with time constraints. Learning javascript frameworks in a backend oriented team.
Accomplishments that we're proud of
We built our first ever deep neural network. We managed to reach our goald of building a working MVP at our first hackaton.
What we learned
We learned how to distribute tasks in a team effectively. We shouldn't eat so much, just because the food is free.
What's next for joke-categorizer
Gathering user feedback on generated categories. Incorporating into our new personal project. Generalization to be useful outside of meme classification.
Built With
- google-cloud
- html
- python
- tensorflow
- vue
Log in or sign up for Devpost to join the conversation.