maderix cb474e1537 Add dynamic weight training pipeline — 110ms/step without recompilation ... Dynamic weight pipeline that eliminates the ~3.7s recompile-every-10-steps bottleneck. Weights are passed via IOSurface spatial dimension instead of baked as constants, so kernels compile once at startup (345ms) and run indefinitely without exec() restart. Key components: - training_dynamic/ — full pipeline (config, IO, MIL generators, train loop) - 9 dynamic kernels shared across all 12 layers - Vocab compaction 32K→9.2K for faster classifier - Vectorized cross-entropy with vDSP/NEON - Adam optimizer with gradient clipping + cosine LR schedule - Checkpoint save/resume - test_dynamic_matmul.m — validates dynamic weight matmul vs cblas - test_weight_patch.m — tests weight update via IOSurface - dashboard.py — updated with --dynamic flag for v2 pipeline support, improved step regex parsing, --scratch/--lr/--accum CLI args Performance: 110ms/step steady-state (no recompile overhead) ane_fwd=21 ane_bwd=28 io_fwd=12 io_bwd=15 silu=10 cls=13 rms=5 ms		2026-03-03 04:34:55 -08:00
..
Makefile	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00
config.h	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00
cpu_ops.h	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00
io.h	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00
mil_dynamic.h	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00
train.m	Add dynamic weight training pipeline — 110ms/step without recompilation	2026-03-03 04:34:55 -08:00