Haoming Song ff7cc3de1d fix(train): narrow the accelerate env guard, and fix a device-bound assert (#4347) ...

Two post-merge CI failures on main, both from #4010.

Benchmark Integration Tests (Libero) — `accelerate launch` exports whole groups
of variables unconditionally (the five ACCELERATE_DYNAMO_* it writes default the
backend to "no"), so matching on prefixes refused launches that configure
nothing, contradicting the documented flow where accelerate is supported as a
plain launcher. The guard now watches only the three switches that hand a
subsystem to the environment.

GPU Tests — `test_metrics_tracker_reduce_across_ranks_invokes_all_reduce`
compared the captured reduction buffer against a CPU tensor, so the assert
raised "Expected all tensors to be on the same device" wherever CUDA is
available. The expected tensor is built on the buffer's device instead.

2026-08-06 18:39:24 +02:00

..

test_accelerate_canaries.py

feat(train): parallel training framework — FSDP2, HSDP, gradient accumulation, and DCP checkpoints (#4010)

2026-08-06 19:16:41 +08:00

test_checkpoint_wrappers.py

feat(train): parallel training framework — FSDP2, HSDP, gradient accumulation, and DCP checkpoints (#4010)

2026-08-06 19:16:41 +08:00

test_multigpu_training.py

feat(train): parallel training framework — FSDP2, HSDP, gradient accumulation, and DCP checkpoints (#4010)

2026-08-06 19:16:41 +08:00

test_parallel_dims_and_factory.py

fix(train): narrow the accelerate env guard, and fix a device-bound assert (#4347)

2026-08-06 18:39:24 +02:00

test_policy_surface.py

feat(train): parallel training framework — FSDP2, HSDP, gradient accumulation, and DCP checkpoints (#4010)

2026-08-06 19:16:41 +08:00