Have you always wanted to build your own neural network, but didn’t know where to start? The course “Neural Networks: Zero to Hero” by Andrej Karpathy will help you.
In this course, you will learn everything you need to understand and implement modern neural networks yourself. Andrej Karpathy will guide you from the basics to advanced concepts – in a very practical way.
What you learn
Backpropagation and training processes: You will understand the basics and apply them yourself.
Language modelling: Build your own bigram model and extend it to a transformation model like GPT.
Multilayer Perceptrons (MLP): Learn how to implement and optimise an MLP.
Activation Functions and Batch Normalisation: Discover advanced techniques to improve your models.
Manual Backpropagation: Perform backpropagation manually step by step and gain deeper insights.
WaveNet: Develop deeper networks with tree-like structures.
GPT Model: Build your own generative pre-trained transformer.
Tokeniser: Create a tokeniser to translate between strings and tokens.
Mit dem Laden des Videos akzeptieren Sie die Datenschutzerklärung von YouTube.
Mehr erfahren
Andrej Karpathy, a leading expert in the field of deep learning, will not only teach you what you need to know, but also show you how to apply it in practice. With detailed explanations and practical examples, you will quickly make progress and be able to create your own neural network.
Basic mathematical knowledge and programming skills in Python are required.
