data-loading-preprocessing
Prepare, inspect, and smoke-test MedicalZooPytorch dataset layouts, medical-image preprocessing, subvolume generation, normalization, resampling, coordinate transforms, and 3D augmentation.
- License 许可
- MIT
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- data-loading-preprocessing
- license
- MIT
- metadata
- { "disco-role": "operating" }
- description
- Prepare, inspect, and smoke-test MedicalZooPytorch dataset layouts, medical-image preprocessing, subvolume generation, normalization, resampling, coordinate transforms, and 3D augmentation.
- disable-model-invocation
- true
02 资源文件 6
- 参考文档references/data-layout.md4.6 KB
- 参考文档references/troubleshooting.md3.8 KB
- 参考文档references/workflows.md4.7 KB
- 脚本scripts/smoke_augmentations.py5.5 KB
- 脚本scripts/smoke_dataloaders.py9.4 KB
- 脚本scripts/smoke_preprocessing.py4.8 KB
03 SKILL.md 正文
Data Loading and Preprocessing
Use this sub-skill for the repository's data-preparation path only:
- dataset folder layouts and manifest formats
- loader arguments and dispatcher entry points
- subvolume generation and cached patch files
- normalization, resampling, cropping, and coordinate transforms
- 3D augmentation operators and synthetic smoke checks
Do not use this route for model selection, training, checkpointing, inference, or loss design. Hand those off to sibling sub-skills.
Start here
references/data-layout.md— dataset folders, filenames, manifests, and generated cache pathsreferences/workflows.md— preprocessing recipes, loader argument map, and smoke-script usagereferences/troubleshooting.md— common missing-data, shape, and transform failuresscripts/smoke_preprocessing.py— synthetic NIfTI preprocessing smokescripts/smoke_augmentations.py— synthetic 3D augmentation smokescripts/smoke_dataloaders.py— synthetic loader and manifest smoke
Key entry points
lib.medloaders.generate_datasets(...)lib.medloaders.select_full_volume_for_infer(...)lib.medloaders.medical_image_process.load_medical_image(...)lib.medloaders.medical_loader_utils.create_sub_volumes(...)lib.augment3D.RandomChoice(...)
Coverage
This route covers the loader and preprocessing families that prepare data for later training or inference:
- ISEG 2017 and 2019
- BraTS 2018, 2019, and 2020
- MRBRAINS 2018
- IXI T1/T2
- MICCAI 2019 Gleason pathology
- COVIDx and COVID CT manifest-based 2D loaders
Native-check note
The bundled scripts use tiny synthetic fixtures only. The repo's original data-loader checks remain blocked until the real datasets are placed in the documented folders.
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).