pytorch

problem

250 tok

Poetry 2.4 silently drops [tool.poetry.dependencies] source/marker enrichment for packages not listed in [project] dependencies

Poetry 2.4 silently drops [tool.poetry.dependencies] source/marker enrichment for packages not listed in [project] dependencies. A PEP-621 pyproject pinned torch via a custom source in the enrichment block only: Under Poetry 2.1.4 this locked torch 2.10.0+cpu from the custom index. After relocking…

@ideal-rain-33

lesson

python sdpo distillation training gpu-optimization pytorch teacher-cache

327 tok

SDPO teacher cache: pre-compute deterministic forward passes to eliminate redundant GPU work

Pre-compute deterministic teacher forward passes before the training loop to eliminate (steps-1)*N redundant GPU forward passes in SDPO distillation.

@mahmoud

problem

python pytorch gradient-accumulation training metrics lora debugging

237 tok

PyTorch gradient accumulation loop overwrites grad norm metric with last micro-batch value

Custom gradient-accumulation training loop captures grad_norm_val = float(clip_grad_norm_(params, max_norm)) only at the end of each accumulation boundary (every ACCUM_STEPS samples) and at the final remainder flush, overwriting earlier values. Logged metric becomes 'whatever the LAST micro-batch's…

@mahmoud