AGENT SKILL

deep_ritz_training_recovery

Run bounded Deep Ritz optimization and emit auditable recovery artifacts for variational PDE experiments.

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

01 前言 Front Matter

name
deep_ritz_training_recovery
description
Run bounded Deep Ritz optimization and emit auditable recovery artifacts for variational PDE experiments.

02 资源文件 3

03 SKILL.md 正文

Deep Ritz Training Recovery

When To Use

Use this skill after residual-network, stochastic-sampling, and variational-loss skills exist. It is for bounded recovery experiments that must prove the Deep Ritz mechanism executed without reading an original repository.

Inputs

  • Attempt directory.
  • Generated skills root.
  • Runtime handoff JSON.
  • Module plan target.
  • Recovery settings: dimension, width, blocks, steps, batch sizes, beta, learning rate, seed.

Outputs

  • recovery/logs/training_trace.json with loss, error, parameter snapshots, and optimizer evidence.
  • recovery/recovery_result.json with metric and mechanism checks.
  • Generated-skill invocation logs.

Workflow

  1. Import or call the generated network, sampler, and loss helpers.
  2. Prefer PyTorch autograd if available.
  3. Run fresh stochastic quadrature samples for each optimizer step.
  4. Record parameter changes with params_before and params_after.
  5. Evaluate relative L2 error on a fixed validation batch.
  6. Mark reduced/proxy recovery honestly when the full paper budget is infeasible.

Validation

Run:

python tests/test_training_recovery.py

Limitations

The default harness is intentionally reduced and bounded. It validates mechanism fidelity, not the paper's full 50,000-step accuracy target.

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