AGENT SKILL

classification

Operate Snorkel PyTorch classification, training, logging, and evaluation APIs.

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

01 前言 Front Matter

name
classification
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Operate Snorkel PyTorch classification, training, logging, and evaluation APIs.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

Classification

Use this sub-skill for Snorkel discriminative modeling workflows built around:

  • DictDataset, DictDataLoader, and collate_dicts
  • Operation, Task, MultitaskClassifier, and cross_entropy_with_probs
  • Trainer, logging, checkpointing, and batch / learning-rate schedulers
  • Scorer, metric_score, label utilities, and error-analysis helpers

Route elsewhere when needed

Start here

  1. Read references/api-reference.md for the public surface and defaults.
  2. Use references/training-and-evaluation.md for the end-to-end workflow.
  3. Check references/troubleshooting.md for common failures and fixes.
  4. Run scripts/classification_smoke.py for a tiny CPU-only validation.

What to emphasize

  • Build the shortest task graph that still exposes the behavior being debugged.
  • Prefer in-memory tensors and small fixtures.
  • Keep label semantics explicit: hard labels for scoring, probabilistic labels for soft-loss training.
  • Use the metric namespace task/dataset/split/metric when reading scores, logs, and checkpoints.

What this sub-skill does not own

  • Weak supervision model training and LF analysis
  • Slice-function authoring and slice-aware modeling details
  • Data augmentation and generic preprocessing pipelines

When a request crosses those boundaries, route it to the adjacent sub-skill instead of stretching this one.

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