mirror of https://github.com/MLSysBook/TinyTorch.git synced 2026-05-01 02:17:33 -05:00

Files

Vijay Janapa Reddi 984772dd40 Add documentation standards and development setup

- Create .claude directory with team structure and guidelines
- Add MODULE_DEVELOPMENT_GUIDELINES.md for educational patterns
- Add EDUCATIONAL_PATTERN_TEMPLATE.md for consistent module structure
- Add GIT_WORKFLOW_STANDARDS.md for branch management
- Create setup-dev.sh for automated environment setup
- Add notebook workflow documentation
- Add CI/CD workflow for notebook testing

This commit establishes consistent development standards and documentation
for the TinyTorch educational ML framework development.

2025-09-15 13:30:10 -04:00

2.1 KiB

Raw Blame History

TinyTorch: Quick Start for Interactive Notebooks

Get up and running with TinyTorch interactive notebooks in under 5 minutes!

🚀 One-Command Setup

# Clone repository
git clone https://github.com/your-org/TinyTorch.git
cd TinyTorch

# Run automated setup
chmod +x setup-dev.sh
./setup-dev.sh

# Activate environment
source .venv/bin/activate

📓 Convert Modules to Notebooks

# Convert all modules to interactive notebooks
python -m tito.main module notebooks

# Or convert specific modules
python -m tito.main module notebooks --module 03_activations

🎯 Start Learning

# Open Jupyter Lab
jupyter lab

# Navigate to modules/source/ and open any .ipynb file
# Try: 03_activations/activations_dev.ipynb

✅ Verify Everything Works

# Run environment check
python -m tito.main system doctor

# Should show:
# ✅ Python 3.8+
# ✅ Virtual Environment Active
# ✅ Essential Dependencies Installed

🆘 Need Help?

Full Documentation: NOTEBOOK_WORKFLOW.md
Environment Issues: python -m tito.main system doctor
Command Help: python -m tito.main --help

📋 Available Modules

After conversion, you'll have interactive notebooks for:

01_setup: Environment setup and course introduction
02_tensor: Tensor operations and broadcasting
03_activations: Neural network activation functions
04_layers: Building neural network layers
05_dense: Fully connected networks
06_spatial: Convolutional neural networks
07_attention: Attention mechanisms and transformers
08_dataloader: Data loading and preprocessing
09_autograd: Automatic differentiation
10_optimizers: Gradient descent and optimization
11_training: Training loops and validation
12_compression: Model compression techniques
13_kernels: Custom CUDA kernels
14_benchmarking: Performance measurement
15_mlops: Production deployment

Each module includes:

📚 Educational content with explanations
💻 Interactive code cells
🧪 Comprehensive tests via NBGrader
🎯 Hands-on exercises and experiments

Happy learning! 🔥

2.1 KiB Raw Blame History