Tagged: Siraj Raval


Simulating a Self-Driving Car

Siraj Raval shows you how to safely experiment with writing self-driving car algorithms and without modifying your kid’s toys. Using Udacity’s car simulator app as an environment to create your own autonomous agent, he shows you how to use Keras to train a convolutional neural network on images from the...


How to Learn from Little Data

These days, the world is awash in data. There’s no shortage of it.  However, sometimes you don’t have all the data you need to glean any insight out of. But all hope is not lost: enter meta-learning. Here’s an interesting look at meta-learning and how to learn from a shortage...


How to Generate Images with Deep Learning

Siraj Raval has a new video out on how to build a variational autoencoder.  It will be capable of generating novel images after being trained on a collection of images. Using handwritten digit images as training data, he’ll both generate new digits and plot out the learned embeddings. He also...