Ageing, content updated Nov 2025
Modern Deep Learning Foundations
Packt on O'Reilly with Dr. Barak Or
A compact tour of deep learning: loss functions, optimisation and regularisation, CNNs, recurrent networks, autoencoders and the Transformer, then mixed precision, fine-tuning, versioning and basic deployment in PyTorch and TensorFlow.
Our take
About 2 hours (Packt, November 2025), so a refresher rather than a full course. Included in an O'Reilly subscription.
Best for: Developers who know basic ML and want a fast, current refresher on deep learning foundations.
Based on the course page, syllabus and pricing, not on completing the course. How we rank
Tools and models covered
- PyTorch
- TensorFlow
- Google Colab
Based on the course content as of 2025-11.
What you'll learn
- Neural networks
- Backpropagation
- CNNs
- RNNs and LSTMs
- Transformers
- Regularization
- Transfer learning
- Model deployment