Working notes from applied deep-learning practice.
September 11, 2026 · LayerGenius
Accuracy on the test set is the last honest number you get for free. What to log at serving time, which drift signals are worth alerting on, and how to decide when a retrain is justified.
Read moreSeptember 10, 2026 · LayerGenius
Data parallelism in Keras 3 across the TensorFlow and JAX backends: global batch size, learning-rate scaling, input-pipeline limits, checkpointing, and the measurements that tell you whether the second GPU earned its keep.
Read moreAugust 6, 2026
One line enables mixed precision — and three known failure modes decide whether it helps. float16 vs bfloat16, loss scaling, the float32 output layer, and the A/B test that settles it.
Read moreJuly 16, 2026
A trained model is not a deliverable — a verified inference artifact is. Export contracts, op coverage, the quantization ladder, and the parity tests that make conversions safe.
Read moreJune 24, 2026
Honest splits, a linear-probe baseline, gradual unfreezing with frozen BatchNorm, domain-respecting augmentation, and one change at a time — the fine-tuning recipe we use in Keras.
Read more