Contents Menu Expand Light mode Dark mode Auto light/dark, in light mode Auto light/dark, in dark mode Skip to content
blox documentation
blox documentation

Getting Started

  • Installation
  • MNIST Classification with blox
  • Recurrent Neural Networks: From Sine Waves to Shakespeare

Key Concepts

  • Design Principles

Advanced Topics

  • Checkpointable Training on Google Colab
  • 1. Setup
  • 2. Configuration
  • 3. Utilities
  • 4. Resumable Dataset
  • 5. Model
  • 6. Training Step
  • 7. Initialize and Restore
  • 8. Training Loop
  • 9. Evaluation
  • 10. Summary
  • Two Approaches
  • 1. Define an MLP Module
  • 2. Explore the Graph
  • 3. The LoRA Pattern
  • 4. Apply LoRA to Selected Layers
  • 5. Initialize LoRA Parameters
  • 6. Freeze Base Weights
  • 7. Training with LoRA
  • 8. Verify Only LoRA Weights Changed
  • 9. Merging LoRA Weights (Optional)
  • Create a LoRA-aware MLP
  • Benefits of the LoRA-aware Approach
  • Toggling LoRA
  • Sharp Bits in blox
  • 1. The Params Container
  • 2. RNG Handling in Parallel Contexts
  • 3. Init vs Runtime
  • 4. Graph and Module Modifications
  • 5. Summary

Reference

  • API Reference
Back to top

Python Module Index

b
 
b
- blox
    blox.blocks
    blox.visualize
Copyright © 2026, Hamza Merzić
Made with Sphinx and @pradyunsg's Furo