Tutorials

Interactive Jupyter notebooks to learn Nabla’s features and capabilities through hands-on examples.

Interactive Notebooks

• Understanding Nabla - Program Transformations (Part 1)
  • 1. Defining and Visualizing a Python Function
  • 3. Gradient Transformation
  • 4. Vectorization Transformation
  • 5. Compilation Transformation with MAX
• Value-and-Grads (CPU)
  • Setup and Imports
  • Define Function with Automatic Differentiation
  • Create Input Tensors
  • Compute Values and Gradients
• Value-and-Grads (GPU)
  • Setup and Imports
  • Define Function with Automatic Differentiation
  • Create Input Tensors
  • Compute Values and Gradients
• MLP Training (CPU)
  • 2. Core Model and Utility Functions
  • 3. Training Step Implementations
  • 4. Experiment Runner
  • 5. Run Experiments and Compare Results
  • 6. Analysis and Conclusions
• MLP Training (GPU)
  • Installation and Setup
  • Introduction to Nabla and GPU Acceleration
  • Summary
• JAX vs. Nabla: Training an MLP (CPU)
  • 2. Core Model Functions (Nabla and JAX)
  • 3. Optimizer Implementations (Nabla and JAX)
  • 4. Training Step Implementations
  • 5. Experiment Runner Functions
  • 6. Run Experiments and Compare Results
• JAX vs. Nabla: Training a Transformer (CPU)
  • 1. Imports and Configuration
  • 2. Positional Encoding
  • 3. Scaled Dot-Product Attention
  • 4. Multi-Head Attention
  • 5. Position-wise Feed-Forward Network
  • 6. Layer Normalization
  • 7. Encoder and Decoder Layers
  • 8. Embedding Lookup
  • 9. Full Transformer Forward Pass
  • 10. Loss Function
  • 11. Parameter Initialization
  • 12. Data Generation
  • 13. Optimizer (AdamW)
  • 14. JIT-Compiled Training Step & Inference
  • 15. JAX Training Run
  • 16. Nabla Training Run
  • 17. Loss Curves Visualization
  • 18. Final Evaluation