sunnydave234 49d5ea49bc fix(utils): add MPS branch to torch RNG state serialization (#4014) ...

serialize_torch_rng_state/deserialize_torch_rng_state only handled CPU
and CUDA generators. On MPS, resumed training was not bit-exact for any
stochastic op (dropout, ACT's CVAE noise) since the MPS generator's state
was never saved or restored. Mirrors the existing CUDA branch using
torch.mps.get_rng_state/set_rng_state (available since torch 2.11).

Note: get_rng_state()/set_rng_state() (used by seeded_context()) have the
same gap but are out of scope here — happy to follow up separately if
useful.

Co-authored-by: Sunny Dave <sunnydave@Sunnys-Mac-Studio.local>
Co-authored-by: Steven Palma <imstevenpmwork@ieee.org>

2026-07-30 12:22:13 +02:00

..

test_collate.py

Add extensive language support (#3467)

2026-05-19 14:46:11 +02:00

test_device_utils.py

fix(utils): raise ValueError from get_safe_torch_device (#3992)

2026-07-29 19:21:50 +02:00

test_encoding_utils.py

chore(utils): move encoding utils and process to their respective modules (#2029)

2025-09-24 16:47:37 +02:00

test_foxglove_visualization.py

feat(visualization): add foxglove support (#3902)

2026-07-01 18:39:32 +02:00

test_hub.py

feat(train): run training remotely on HF Jobs via --job.target (#3856)

2026-06-29 17:59:33 +02:00

test_io_utils.py

chore(docs): add missing license headers (#2140)

2025-10-08 14:27:52 +02:00

test_keyboard_input.py

feat(utils): display-independent keyboard controls for recording (Wayland / headless / macOS) (#3875)

2026-06-25 10:58:39 +02:00

test_logging_utils.py

Aggregate policy sub-losses through MetricsTracker (#4024)

2026-07-16 12:12:37 +02:00

test_process.py

RL stack refactoring (#3075)

2026-05-12 15:49:54 +02:00

test_random_utils.py

fix(utils): add MPS branch to torch RNG state serialization (#4014)

2026-07-30 12:22:13 +02:00

test_replay_buffer.py

RL stack refactoring (#3075)

2026-05-12 15:49:54 +02:00

test_rerun_visualization.py

fix(depth unit): adding input depth unit storage in the dataset metadata (#3899)

2026-07-02 11:53:13 +02:00

test_rotation.py

fix(utils): reject zero-norm / invalid quaternions in Rotation (#3988)

2026-07-29 19:13:40 +02:00

test_sample_weighting.py

Reward models refactor (#3142)

2026-04-28 17:56:24 +02:00

test_train_utils.py

feat(train): run training remotely on HF Jobs via --job.target (#3856)

2026-06-29 17:59:33 +02:00

test_visualization_utils.py

feat(visualization): add foxglove support (#3902)

2026-07-01 18:39:32 +02:00