What it does
Shape is a machine learning, realtime cloud database updating, and most importantly, realistic and efficient app for health and fitness.
How we built it
Challenges we ran into
The Watson API was difficult to use at times because many times it would reject classification of a jpg file and at other times, the slightest bit of a syntax error could prevent progress for hours. Also, obtaining the sheer amount of data needed to train the custom classifier we built proved to be very difficult. Over 1600 image urls were obtained throughout the process of this hackathon, and not all of them were able to be used as training data for the classifier as we were limited by the capacity of the free version of the Visual Learning Watson API.
Accomplishments that we're proud of
One of the accomplishments that we are very proud of is our ability to learn new technologies very quickly. All of the technologies that we used were new to everyone in the group. None of us knew how to use PhoneGap. None of us knew how to use Watson’s Image Visualization. None of us knew how to use Google Firebase. Despite all of that, we helped each other learn these new technologies, and we combined them together to make a useful, functional app.
Another accomplishment that we are proud of is our ability to adapt to present challenges and obstacles.