Every empirical statement in the main text relies on the companion notebooks listed in the Execution Map. Bit-exact reproducibility on a different machine requires fixing both the random seeds and the floating-point environment; Table E.1 summarizes the conventions used in this script’s notebooks.
Table E.1:Reproducibility conventions used by the companion notebooks. The conventions pin random seeds and document hardware, software, and precision choices; bit-exact reproduction additionally requires the deterministic settings and caveats described below the table.
| Item | Convention |
|---|---|
| Random seeds | Each notebook declares a SEED constant in cell 1 (default 0); the corresponding framework seeds (numpy.random.seed(SEED), torch.manual_seed(SEED), tf.random.set_seed(SEED), and jax.random.PRNGKey(SEED) where the framework is used) are derived from it. Per-cell offsets and per-iteration deviations are documented inline. Notebook-by-notebook normalization against this convention is tracked through the chapter audits; some legacy notebooks still hard-code a literal seed and are scheduled for normalization. |
| Run-mode budget | Most notebooks declare RUN_MODE {smoke, teaching, production} alongside SEED in cell 1. smoke is sized for a single CPU core (CI-friendly); teaching for one consumer GPU; production for an A100 or larger. Per-notebook hyperparameters (epochs, batch sizes, restart counts) are gated on RUN_MODE. |
| Hardware | Classroom-scale runs target a single CPU core or one consumer GPU (e.g. NVIDIA T4 / RTX 3060). Production runs use one A100 or larger; this is documented per-notebook in the chapter. |
| Software stack | Python , TensorFlow , PyTorch , JAX , GPyTorch . Minimum versions are listed in requirements.txt at the root of the companion repository. JAX appears in the Krusell–Smith warm-up notebook in lectures/lecture_10_* (sequence-space DEQNs); the other lectures use TensorFlow or PyTorch. |
| Numerical precision | TensorFlow / PyTorch default to float32; a number of notebooks switch to float64 where second-order derivatives or long simulations make the last digits matter (documented inline). |
| GPU determinism | Optional, bit-exact runs only: set CUBLAS_WORKSPACE_CONFIG=:4096:8 and torch.use_deterministic_algorithms(True) on the PyTorch side, and call tf.config.experimental.enable_op_determinism() on the TensorFlow side. Default classroom and production runs leave these unset, accepting last-decimal nondeterminism in exchange for speed; expect a small slowdown when the flags are on. |
| Reported numbers | Where the script states an accuracy or runtime, the source is either the cited paper (for production-scale results) or the companion notebook with the seed above (for classroom-scale results). |
For full bit-exact reproduction, e.g. for a regulatory-grade audit, the determinism flags above are necessary but not sufficient: GPU non-determinism in atomic accumulators and BLAS implementation differences across CUDA versions can still cause diverging results in the last few decimal places. This is a known limitation of GPU-accelerated deep learning and is one of the reasons Section 12.6 flags reproducibility-critical settings as a regime where deterministic finite-difference solvers may still be preferred.