data-preparation
Prepare and validate PaddleGAN dataset layouts, dataroot mappings, and preprocessing inputs for translation, super-resolution, and lip-sync workflows.
- License 许可
- Apache-2.0
- 最近更新
- 2026-10-08
- Stars
- ★ 11
- Forks
- 3
01 前言 Front Matter
- name
- data-preparation
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Prepare and validate PaddleGAN dataset layouts, dataroot mappings, and preprocessing inputs for translation, super-resolution, and lip-sync workflows.
- disable-model-invocation
- true
02 资源文件 3
- 参考文档references/dataset-layouts.md4.9 KB
- 参考文档references/preprocessing-scripts.md3.2 KB
- 参考文档references/troubleshooting.md4.2 KB
03 SKILL.md 正文
Data Preparation
Use this sub-skill when the task is about making PaddleGAN data usable before training or inference.
It covers:
- CycleGAN / U-GAT-IT style unpaired layouts
- Pix2Pix-style paired layouts with split AB images
- custom paired or unpaired image folders
- DIV2K cropping and other SR inputs
- LRS2 preprocessing for Wav2Lip
- REDS and Vimeo90K sequence expectations
- RealSR synthetic degradation outputs
- config
datarootand related path updates
Entry points
references/dataset-layouts.mdreferences/preprocessing-scripts.mdreferences/troubleshooting.md
Bundled helpers
scripts/check_dataset_layout.pyscripts/process_div2k_data.py
Route elsewhere
- train/evaluate/resume command construction →
training-configs - export/static inference/deployment outputs →
deployment-export - benchmark execution and performance harnesses → out of scope
Guardrails
- Do not run downloads, training, inference, or face/video preprocessing by default.
- Prefer explicit local paths over hidden repository defaults.
- If a dataset layout is ambiguous, validate the directory tree first with
scripts/check_dataset_layout.py.
04 同仓库更多 Skills
obsidian-clipper-template-creator
Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.
obsidian-clipper-template-creator
Guide for creating templates for the Obsidian Web Clipper. Use when you want to create a new clipping template, understand available variables, or format clipped content.
python-api
Programmatic Object-Detection-Metrics API guidance for BoundingBox, BoundingBoxes, Evaluator, enums, coordinate conversion, AP/mAP semantics, and source-style import diagnostics.
file-evaluation
Folder-based PASCAL VOC AP/mAP evaluation from ground-truth and detection text files.
object-detection-metrics
Guides legacy Object-Detection-Metrics PASCAL VOC AP/mAP evaluation from detection text folders and direct Python API classes.
nx-workspace-patterns
Configure and optimize Nx monorepo workspaces. Use when setting up Nx, configuring project boundaries, optimizing build caching, or implementing affected commands.