👩💻 Join our community of thousands of amazing developers!
Before I started learning about AI, I thought that training a neural network meant training from scratch. Want to train a network to recognize dogs? Feed it 10,000 dogs and watch it go. Want to recognize images of malignant tumors in a CT scan? Collect a bunch of medical images and train away. I didn’t realize you could take the models trained on one set of images and apply them to another, entirely unrelated set. Turns out this is exactly what you can do! I’ve been working through the fast.ai c...