AGENT SKILL

training-loop-integration

Migrate raw PyTorch training and evaluation loops to Hugging Face Accelerate using Accelerator, prepare(), backward(), gradient accumulation, dataloader behavior, gather/reduce, mixed precision, DDP kwargs, local SGD, communication hooks, and basic distributed logging.

License 许可
Apache-2.0
最近更新
2026-09-03
Stars
★ 351
Forks
27

01 前言 Front Matter

name
training-loop-integration
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Migrate raw PyTorch training and evaluation loops to Hugging Face Accelerate using Accelerator, prepare(), backward(), gradient accumulation, dataloader behavior, gather/reduce, mixed precision, DDP kwargs, local SGD, communication hooks, and basic distributed logging.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Training Loop Integration

Use this sub-skill when adapting an existing PyTorch model, optimizer, dataloader, scheduler, or evaluation loop to Accelerate while keeping the training code mostly framework-neutral.

Route first

  • For accelerate config, accelerate launch, launch flags, config files, or process-count setup, use ../configuration-and-cli/.
  • For DeepSpeed, FSDP, TPU, Megatron-LM, FP8 backend recipes, or backend-specific plugin tuning, use ../distributed-training-backends/.
  • For save_state, load_state, model checkpoint saving, trackers, profiler, or experiment dashboards, use ../checkpointing-and-tracking/.
  • For large-model loading, device maps, offload, or inference dispatch, use the root skill routing to the big-model sub-skill.

Start here

  1. Read references/workflows.md for migration recipes and loop patterns.
  2. Read references/api-reference.md for constructor options, wrappers, dataloader semantics, gather/reduce, kwargs handlers, logging, and scheduler behavior.
  3. Read references/troubleshooting.md when a distributed loop hangs, metrics are wrong, gradients do not sync, or device placement behaves unexpectedly.
  4. Run scripts/accelerator_loop_smoke.py to verify a minimal CPU training loop in the current environment.

Core migration checklist

  • Create one Accelerator early, before distributed-sensitive objects need its state.
  • Remove hard-coded .cuda() and direct .to("cuda"); prefer automatic device placement through accelerator.prepare().
  • Pass related PyTorch objects to accelerator.prepare(...) and unpack them in the same order.
  • Replace loss.backward() with accelerator.backward(loss).
  • For accumulation, create Accelerator(gradient_accumulation_steps=N) or a GradientAccumulationPlugin, then wrap each minibatch body in with accelerator.accumulate(model):.
  • Use accelerator.gather_for_metrics(...) for evaluation metrics and accelerator.pad_across_processes(...) before gathering variable-size tensors.
  • Use accelerator.print(...) or accelerate.logging.get_logger(...) instead of unsynchronized process-wide print() spam.

Smoke check

python skills/accelerate/sub-skills/training-loop-integration/scripts/accelerator_loop_smoke.py
python skills/accelerate/sub-skills/training-loop-integration/scripts/accelerator_loop_smoke.py --help

The smoke script is intentionally CPU-safe and tiny. It validates importability, Accelerator.prepare, Accelerator.accumulate, Accelerator.backward, scheduler wrapping, and metric gathering without requiring a distributed launcher.

04 同仓库更多 Skills

VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-triton-opt

Make a single PassNet Triton kernel fast AND numerically faithful: performance model (when a replacement can win at all), block/grid/warp tuning, autotune policy, launch overhead, and per-op numeric recipes to pass the dtype baseline tolerances. Use when a pass matches and is correct but speedup below expected, or when correctness fails by small numeric margins.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-solve

END-TO-END playbook for solving one PassNet sample: analyze the computation graphs, decide the optimization strategy (what to fuse, what to replace, what to leave alone), drive the iteration loop, and maximize the sample score. This is the ENTRY skill — invoke it first for any PassNet optimization task; it tells you when to use passnet-pattern-fusion, passnet-triton-opt and passnet-feedback.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-skill

PassNet GPU kernel optimization via compiler passes. Design and implement Triton-based optimization passes, create pass files under ./pass_dir/, self-evaluate with pass_evaluator, and iterate to maximize GPU speedup.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-pattern-fusion

Author PassNet pass files: write patterns that actually MATCH the FX graph, pick fusion regions (which ops to absorb into one Triton kernel), and structure multi-pass submissions with the mandatory shared-dispatch architecture. Use when creating or fixing pattern()/replacement_args()/replacement_func() files, when a pass "failed to match", or when deciding how to fuse multiple kernels/ops.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-orchestrate

Round/batch planning and decision support for PassNet graph optimization. Use for multi-sample triage, worker allocation, eval-budget policy, and post-eval keep/revert decisions. For solving one concrete sample end to end, use passnet-solve as the entry skill; consult this skill only when you need broader planning or a specific decision gate.

★ 351
VectorSpaceLab
VectorSpaceLab/AREX-Skill

passnet-feedback

Fast, deterministic feedback for PassNet work: pre-flight pattern/match verification WITHOUT burning a GPU evaluation, per-node bottleneck analysis of a sample's graphs, and eval-log parsing into per-variant status + estimated score + failure classification. Use BEFORE every GPU evaluation (check_pattern), at the START of a sample (analyze_graph), and AFTER every evaluation (parse_eval_log).

★ 351