mirror of
https://github.com/MLSysBook/TinyTorch.git
synced 2026-05-06 04:27:30 -05:00
f9309e8b9d
📦 Module File Organization: - Renamed networks_dev.py → dense_dev.py in 05_dense module - Renamed cnn_dev.py → spatial_dev.py in 06_spatial module - Added new 07_attention module with attention_dev.py - Updated module.yaml files to reference correct filenames - Updated #| default_exp directives for proper package exports 🔄 Core Package Updates: - Added tinytorch.core.dense (Sequential, MLP architectures) - Added tinytorch.core.spatial (Conv2D, pooling operations) - Added tinytorch.core.attention (self-attention mechanisms) - Updated all core modules with latest implementations - Fixed tensor assignment issues in compression module 🧪 Test Integration Fixes: - Updated integration tests to use correct module imports - Fixed tensor activation tests for new module structure - Ensured compatibility with renamed components - Maintained 100% individual module test success rate Result: Complete 14-module TinyTorch framework with proper organization, working integrations, and comprehensive test coverage ready for production use.
🧪 TinyTorch Integration Tests
⚠️ CRITICAL DIRECTORY - DO NOT DELETE
This directory contains 17 integration test files that verify cross-module functionality across the entire TinyTorch system. These tests represent significant development effort and are essential for:
- Module integration validation
- Cross-component compatibility
- Real-world ML pipeline testing
- System-level regression detection
📁 Test Structure
test_*_integration.py- Cross-module integration teststest_utils.py- Shared testing utilitiestest_integration_report.md- Test documentation
🧪 Integration Test Coverage
Foundation Integration
test_tensor_activations_integration.py- Tensor + Activationstest_layers_networks_integration.py- Layers + Dense Networkstest_tensor_autograd_integration.py- Tensor + Autograd
Architecture Integration
test_tensor_attention_integration.py- NEW: Tensor + Attention mechanismstest_attention_pipeline_integration.py- NEW: Complete transformer-like pipelinestest_tensor_cnn_integration.py- Tensor + Spatial/CNNtest_cnn_networks_integration.py- Spatial + Dense Networkstest_cnn_pipeline_integration.py- Complete CNN pipelines
Training & Data Integration
test_dataloader_tensor_integration.py- DataLoader + Tensortest_training_integration.py- Complete training workflowstest_ml_pipeline_integration.py- End-to-end ML pipelines
Inference Serving Integration
test_compression_integration.py- Model compressiontest_kernels_integration.py- Custom operationstest_benchmarking_integration.py- Performance measurementtest_mlops_integration.py- Deployment and serving
🔧 Usage
# Run all integration tests
pytest tests/ -v
# Run specific module integration
pytest tests/test_tensor_attention_integration.py -v
pytest tests/test_attention_pipeline_integration.py -v
# Run attention-related tests
pytest tests/ -k "attention" -v
🚨 Recovery Instructions
If accidentally deleted:
git checkout HEAD -- tests/
git status # Verify recovery
📊 Test Coverage
These integration tests complement the inline tests in each module's *_dev.py files, providing comprehensive system validation with focus on:
- Real component integration (not mocks)
- Cross-module compatibility
- Realistic ML workflows (classification, seq2seq, transformers)
- Performance and scalability