In this discussion we explored how we might build decoupled nets.
A discussion on data normalisation and pose estimation
This Thursday I'll show you data augmentation, which I quite like. A little bit of cropping and flipping can make a model think a cat is an airplane. But a little fine tuning can put a stop to that. I'll show you how you can train your model to handle more varied input data and… Continue reading Data augmentation
Are you interested in deep learning and machine learning? Do you prefer to understand by programming?
We highly recommend working through the 'Getting Started' notes from the fast.ai course "Practical Deep Learning for Coders"