mirror of
https://github.com/MLSysBook/TinyTorch.git
synced 2026-05-26 09:57:31 -05:00
- Removed Section 9: Systems Analysis (118 lines) - Removed analyze_compression_accuracy_tradeoff function (56 lines) - Replaced minimal Tensor/Linear implementations with proper imports (57 lines saved) - Added CompressionComplete export class with all core methods (120 lines) - Net reduction: 111 lines (7%) Result: 1564 → 1453 lines Focus: Core compression techniques (pruning, distillation, low-rank) Imports: Now uses tinytorch.core.tensor and tinytorch.core.layers