Build neural networks from scratch. Master CNNs for images, RNNs for sequences, and Transformers using the research-standard PyTorch library.
Deep Learning powers the current AI boom. This course takes you from the basics of a single neuron to building complex architectures used in research and industry. You will master PyTorch tensors and the autograd engine to build neural networks from the ground up. Dive deep into Convolutional Neural Networks (CNNs) for computer vision and Recurrent Neural Networks (RNNs) for sequential data. We concludes with an introduction to Transformers, the architecture behind GPT and BERT.
Estimated completion time: 21 lessons • Self-paced learning • Lifetime access
PyTorch is preferred for research and flexibility.
We use Google Colab (free cloud GPUs).
Linear algebra and calculus are very helpful.
It teaches the architecture (Transformers) behind GenAI.
3 recommended paths based on what you're learning
AI/ML Engineering Lead is the career move most Deep Learning with PyTorch learners don't see coming.
People who combine Deep Learning with PyTorch with Data Ethics tend to stand out. Here's how.
Hugging Face AutoTrain can fine-tune models without writing code. It's like having an assistant on speed dial.