jgrusewski
216db0301d
fix(gpu): eliminate all GPU→CPU roundtrip violations — zero guard findings
Replace .to_vec1()/.to_vec2() bulk downloads with GPU-resident ops:
- PPO/DQN action selection: Gumbel-max trick (categorical on GPU)
- Scalar readbacks: .to_scalar() instead of .to_vec1()[0]
- GPU stats: abs().max(), sqr().sum_all() — single scalar out
- NaN/Inf check: sum_all().to_scalar().is_finite()
- Guard exclusions: inference output boundaries + CPU fallback with GPU path
26 files across ml-ppo, ml-dqn, ml-supervised, ml (ensemble adapters, metrics, data_loading)
Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
2026-03-16 00:19:09 +01:00
..
2026-03-12 01:44:03 +01:00
2026-03-13 11:56:56 +01:00
2026-03-12 01:44:03 +01:00
2026-02-25 23:20:44 +01:00
2026-03-04 15:29:15 +01:00
2026-03-01 23:41:13 +01:00
2026-03-04 23:46:36 +01:00
2026-02-25 23:24:53 +01:00
2026-03-04 23:46:36 +01:00
2026-02-26 09:04:58 +01:00
2026-03-16 00:19:09 +01:00
2026-03-11 22:00:13 +01:00
2026-02-26 00:50:25 +01:00
2026-02-26 00:50:25 +01:00
2026-03-11 22:00:13 +01:00
2026-02-26 00:50:25 +01:00
2026-03-15 11:58:40 +01:00
2026-03-13 11:54:22 +01:00
2026-03-04 22:08:31 +01:00
2026-03-04 22:08:31 +01:00
2026-03-04 22:08:31 +01:00
2026-03-04 22:08:31 +01:00