conditioning
Use diffusion wrapper models conditioned on lower-rate waveforms, mel spectrograms, latents, and transform plugins.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- conditioning
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Use diffusion wrapper models conditioned on lower-rate waveforms, mel spectrograms, latents, and transform plugins.
- disable-model-invocation
- true
02 资源文件 4
- 参考文档references/api-reference.md6.9 KB
- 参考文档references/troubleshooting.md3.5 KB
- 参考文档references/workflows.md4.2 KB
- 脚本scripts/tiny_conditioning_smoke.py5.6 KB
03 SKILL.md 正文
Conditioning
Use this sub-skill when the task is about diffusion wrappers that condition generation on another audio representation.
Route here when the request mentions
- audio upsampling or super-resolution
- waveform to mel vocoding
- diffusion autoencoding or latent conditioning
- encoder, adapter, mel, or transform wrapper setup
AppendChannelsPluginorLTPlugin
Route to ../generation instead when the task is about
- base unconditional generation
- text-conditioned generation
- inpainting
- core UNet, diffusion, or sampler configuration that is not wrapper-specific
What to open first
references/api-reference.mdfor constructors, prefixes, and shape contractsreferences/workflows.mdfor upsampler, vocoder, and autoencoder recipesreferences/troubleshooting.mdfor prefix, shape, and dependency fixesscripts/tiny_conditioning_smoke.pyfor a safe CPU smoke run
Quick map
- Upsampler:
DiffusionUpsampler - Vocoder:
DiffusionVocoder - Autoencoder:
DiffusionAE - Encoder contract:
EncoderBase - Adapter hooks:
AdapterBase - Mel helper:
MelSpectrogram - Transform helpers:
AppendChannelsPlugin,LTPlugin
The bundled smoke uses a local dummy encoder and stays away from optional external encoder packages.
04 同仓库更多 Skills
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.
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.
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.
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.
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.
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).