mirror of
https://github.com/MLSysBook/TinyTorch.git
synced 2026-06-01 21:41:13 -05:00
bf6ba7233c
- Delete 09_config module (redundant with tinytorch/configs/) - Split training into two focused modules: - 08_optimizers: SGD, Adam, learning rate scheduling (core algorithms) - 09_training: Training loops, metrics, checkpointing (orchestration) - Shift subsequent modules down by one number - Better separation of concerns: algorithms vs. systems coordination - Follows PyTorch structure: torch.optim vs training scripts Final progression (00-16): 00_setup → 01_tensor → 02_activations → 03_layers → 04_networks → 05_dataloader → 06_cnn → 07_autograd → 08_optimizers → 09_training → 10_profiling → 11_compression → 12_kernels → 13_benchmarking → 14_mlops → 15_transformer