AGENT SKILL

inference-evaluation

Run nnU-Net v2 prediction from dataset/results or explicit model folders, cascades, TTA/tiling/device/process controls, ensembling, postprocessing, evaluation, best-configuration selection, model sharing, and inference troubleshooting.

License 许可
Apache-2.0
最近更新
2026-10-08
Stars
★ 11
Forks
3

01 前言 Front Matter

name
inference-evaluation
license
Apache 2.0
metadata
{ "disco-role": "operating" }
description
Run nnU-Net v2 prediction from dataset/results or explicit model folders, cascades, TTA/tiling/device/process controls, ensembling, postprocessing, evaluation, best-configuration selection, model sharing, and inference troubleshooting.
disable-model-invocation
true

02 资源文件 4

03 SKILL.md 正文

nnU-Net v2 Inference and Evaluation

Use this sub-skill after a trained, installed, exported, downloaded, or otherwise available nnU-Net v2 model exists and the task is to generate predictions, compare configurations, ensemble outputs, evaluate segmentations, apply postprocessing, or move models between environments.

Route here for

  • Running nnUNetv2_predict from nnUNet_results with dataset id/name, trainer, plans, configuration, folds, checkpoint, device, TTA, tiling, process, and split-batch flags.
  • Running nnUNetv2_predict_from_modelfolder when the model folder is explicit or nnUNet_results is unavailable.
  • Providing previous-stage predictions for cascaded models with -prev_stage_predictions.
  • Saving probability maps for downstream nnUNetv2_ensemble and applying postprocessing.pkl.
  • Evaluating predictions with nnUNetv2_evaluate_folder or nnUNetv2_evaluate_simple, including ignore labels and partial-prediction --chill behavior.
  • Selecting the best trained configuration with nnUNetv2_find_best_configuration after validation probabilities exist.
  • Exporting, installing, downloading, or sharing trained/pretrained models, including custom-trainer availability requirements.

Do not handle here

  • Creating dataset folders, channel names, labels, or inference input naming from scratch: use data-preparation.
  • Planning, preprocessing, experiment planners, or fingerprints: use planning-preprocessing.
  • Training folds, validation reruns, --npz recovery, checkpoint creation, or trainer selection for training: use training-configuration.
  • Implementing custom trainer subclasses or extension code: use customization-extension; this sub-skill only explains how inference locates custom trainer classes.

Fast Path

  1. Confirm the model exists and matches the input dataset: trained result folder with dataset.json, plans.json, and requested fold_* checkpoints, or an explicit model folder with the same files.
  2. Check inference image filenames before running expensive prediction:
    python sub-skills/inference-evaluation/scripts/check_inference_inputs.py INPUT_FOLDER MODEL_OR_DATASET_JSON
    
  3. Use nnUNetv2_predict when the model is in nnUNet_results:
    nnUNetv2_predict -i INPUT_FOLDER -o OUTPUT_FOLDER -d DATASET001_Example -c 3d_fullres -f 0 1 2 3 4
    
  4. Use nnUNetv2_predict_from_modelfolder when the model folder is explicit:
    nnUNetv2_predict_from_modelfolder -i INPUT_FOLDER -o OUTPUT_FOLDER -m MODEL_FOLDER -f 0 1 2 3 4
    
  5. Add --save_probabilities when predictions will be ensembled; run nnUNetv2_ensemble, then apply postprocessing if a postprocessing.pkl was selected.
  6. Evaluate predictions with nnUNetv2_evaluate_folder when dataset.json and plans.json are available, or nnUNetv2_evaluate_simple for label-list evaluation.

Bundled References

  • references/cli-reference.md: exact command routes, flags, defaults, and entry-point distinctions.
  • references/inference-and-evaluation.md: workflows for prediction, cascades, ensembling, postprocessing, evaluation, best-configuration selection, and Python API use.
  • references/model-sharing.md: export/import/download commands and custom-trainer portability rules.
  • references/troubleshooting.md: failure-mode triage for input names, checkpoints, model folders, probabilities, postprocessing metadata, split jobs, devices, and custom trainers.

Bundled Helper

check_inference_inputs.py validates input-folder case/channel naming against dataset.json without loading images or importing nnU-Net. It accepts either a dataset.json path or a model/results folder containing dataset.json.

04 同仓库更多 Skills

lza6
lza6/Armor-piercing-system-prompts

ci-cd-and-automation

Automates CI/CD pipeline setup. Use when setting up or modifying build and deployment pipelines. Use when you need to automate quality gates, configure test runners in CI, or establish deployment strategies.

★ 11
lza6
lza6/Armor-piercing-system-prompts

chrome-devtools

Uses Chrome DevTools via MCP for efficient debugging, troubleshooting and browser automation. Use when debugging web pages, automating browser interactions, analyzing performance, or inspecting network requests. This skill does not apply to `--slim` mode (MCP configuration).

★ 11
lza6
lza6/Armor-piercing-system-prompts

chrome-cdp

Drive a real Chrome via the Chrome DevTools Protocol (CDP). Ensures a debuggable Chrome is running (launching it if needed) so the agent can navigate, read page content, inspect the DOM, run JavaScript, and take screenshots — without the user typing any launch command. Includes a first-class "review a live running app" flow: point Chrome at the app under development and inspect/screenshot the REAL page (what a static-artifact tool cannot do). Backed by Glimpse and the chrome-devtools MCP server. Use when asked to "use chrome", "open chrome", "drive my browser", "read the page in chrome", "review the running app", "screenshot the app", or "/chrome-cdp".

★ 11
lza6
lza6/Armor-piercing-system-prompts

check-identity-pack

Run an AFP 100-point or AUSTRAC safe-harbour identity check over a set of documents, and report exactly what's missing. Use when the user asks to check identity documents, verify someone's ID for onboarding, or assess whether a document pack satisfies Australian identity requirements.

★ 11
lza6
lza6/Armor-piercing-system-prompts

changelog

Auto-generates a changelog from git commits, sprint data, and design documents. Produces both internal and player-facing versions.

★ 11
lza6
lza6/Armor-piercing-system-prompts

cellxgene-census

Query the CZ CELLxGENE Census programmatically for versioned public single-cell and spatial transcriptomics data. Use when you need population-scale cell metadata, gene expression slices, Census summary counts, source H5AD URIs/downloads, embeddings, spatial Census data, or reference atlas comparisons across organisms, tissues, diseases, assays, and cell types. For analyzing your own local single-cell data use scanpy, anndata, or scvi-tools.

★ 11