Working notes from applied deep-learning practice.
September 16, 2026 · LayerGenius
Before you fine-tune a pretrained language model, price the alternatives. Then: LoRA ranks, sequence length, bfloat16, memory math, and how to serve an adapter you can actually roll back.
Read moreSeptember 15, 2026
Most time-series projects fail on evaluation, not architecture. Seasonal-naive baselines, leakage-safe windowing with tf.data, rolling-origin backtesting, and the small set of Keras models worth trying.
Read moreSeptember 14, 2026 · LayerGenius
The server-side half of deployment: exporting a clean serving signature, numbered version directories and rollback, server-side batching for throughput, and how to measure p99 latency before you promise it.
Read moreSeptember 14, 2026
Bit-identical reruns are a tooling problem, not a research one. Seeding across backends, GPU determinism and what it costs, pinning the data pipeline, and the run record that makes a result defensible six months later.
Read moreSeptember 14, 2026
Keras 3 runs the same model on three backends, but the right one is a measurement, not an opinion. A harness that compares step time, memory, and convergence honestly — plus the constraints that usually decide it before the numbers do.
Read more