First project of the second term in Udacity's Robotics Nanodegree. It was about training a neural network to do inference on specific hand gestures.
The project involved:
- Making my own dataset
- Performing data augmentation to make more data to train on
- Performing experiments to choose the suitable network architecture
- Using nvidia digits to monitor hte training process
The network model is available here.
I uplodaed it separately because it was too large to upload on github.