unsloth/studio/backend/core/training
LeoBorcherding c55aaa5809 fix: detect BNB ROCm DLL suffix dynamically instead of hardcoding '72'
BNB_ROCM_VERSION was pinned to '72' which works today (AMD wheel ships
rocm72.dll) but would break again if AMD ships a future wheel with a
different DLL suffix (e.g. rocm713.dll).

Add _detect_bnb_rocm_dll_ver() to install_python_stack.py: scans the
installed bitsandbytes package dir for libbitsandbytes_rocm{VER}.dll
using importlib.util.find_spec (no BNB import needed) and returns the
suffix.  '72' remains the fallback when detection fails.

Apply the same detection inline in worker.py section 1f.  Both paths
still respect a pre-set BNB_ROCM_VERSION (caller override wins).

Tests: +8 cases covering detection logic and fallback (147 passed, 2 skipped).
2026-05-15 14:05:33 -05:00
..
__init__.py Final cleanup 2026-03-12 18:28:04 +00:00
resume.py Studio: Add checkpoint resume for stopped training runs (#5255) 2026-05-04 00:34:46 +04:00
trainer.py Studio: Passing batch size for eval (#5168) 2026-05-14 17:48:28 +04:00
training.py studio: drop unused max_grad_value schema + route plumbing (#5424) 2026-05-14 05:43:58 -07:00
worker.py fix: detect BNB ROCm DLL suffix dynamically instead of hardcoding '72' 2026-05-15 14:05:33 -05:00