The steps we took to make our AI
We learned Pybasics to help us make our Ai.
We then decided on a project and collected 1000 pieces of data and labeled the images. .
Feed the images to yolo the images to train our AI to be able to detect the difference between the types of data.
We then used applications called Flask and nice pages to construct our website and AI in a much neater fashion.
To improve our Ai we could Have our Ai larn from more pictures(right now we only have around 1000 pictures to train)
We could give the AI more time to configure the files so it has more time to detect what is in the image, allowing for more accurate detection.
Another way to improve our AI is by separating all the dwarf planets into separate files so that we could allow our Ai to better detect the different dwarf planets rather than just detecting a dwarf planet.
Our Ai is not 100% accurate because there was a lot of confusion between the properties of a moon and the properties of an asteroid.
You use our Ai by taking or finding a picture of a planet, asteroid, dwarf planet, or moon and submit it to our AI. it will then look at your image and put it into the category it belongs in.
It took us 1-2 days to train our AI and have it learn from 1000 images, it will take longer depending on how much data you need to teach it.