Inspiration
I needed to not have a failed project like last year. So I thought I would try to do a simple project.
What it does
It (inaccurately 36%) classifies images if they are a animal, plant, relic, food, or human.
How I built it
I trained a ai on 5000 images of humans, animals, plants, food, and relics.
Challenges I ran into
Data sorting. Making sure that the images were not corrupted or were too abstract.
Accomplishments that I'm proud of
Creating something that outputs a vector with no NaN floats.
What I learned
Training ai takes more time than expected.
What's next for Lightweight image classifier.
More training. Lots and lots of training.
Built With
- keras
- python
Log in or sign up for Devpost to join the conversation.