On Saturday 27, 2019, at 10:17 AM - Sunday 28, 2019 at 8:45 AM, at LaunchHacks '19, Team Binary was hard at work.

During the process of trying to find a plausible and helpful idea which we would work on for the next 24 hours and we realized that with the recent advances in Machine Learning we can use TensorFlow and Python to solve major problems in just 24 hours. After brainstorming more for a while, we decided to attempt.

What it does

Machine Learning Platform can:

  • Detect cancer
  • Teach people AI + ML
  • Save LIVES!

How we built it

    • Used for styling
    • Used for JavaScript
  • JINJA 2
    • Used for the HTML, so we could use template binding
    • A python web server that is easy to deploy
  • PYTHON 3
    • The Machine Learning/Artificial Intelligence Logic.
  • GIT
    • To track progress.
  • JSON
    • Files that store a model's information.
    • The documentation's language.
  • CSV
    • CSV(Comma Separated Values) files store input and output data
  • .H5 FILES
    • Pickle files
    • Configure Flask
    • A lot of hard work!

Challenges we ran into

During the production of Machine Learning Platform, we had a lot of Python 3 dependencies issue along with Flask issues. It took us about 30 minutes to resolve an issue.

Accomplishments that we're proud of

We're proud of Machine Learning Platform, and how it predicts cancer so accurately.

What we learned

Throughout this hackathon and project, we all learned a lot. We learned about PyTorch, TensorFlow, Naive Efficient Algorithms, and Flask. Now we have this knowledge, we will go henceforward to apply our newfound knowledge, and build something that the world will LOVE.

What's next for Machine Learning Platform

Our project was just a starting point. We would like to add a way for users to upload images(for skin cancer), treat all the cancers and diseases in the world, and teach people Artificial Intelligence and Machine Learning as if someone already knows something, why not share it? Your idea can only make an idea better and better and better...

Share this project: