القائمة
Use when the user's dataset format differs from what the target repo's data loaders expect, to adapt the code (not the data) for compatibility.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | hpc-dataset-adaptation |
| description | Use when the user's dataset format differs from what the target repo's data loaders expect, to adapt the code (not the data) for compatibility. |
Use when the target repo's loader does not read the user's dataset directly. Adapt code to read the data as-is; do not bulk-convert or copy large datasets unless the user explicitly asks.
Skip this if the data already matches the repo's expected format.
Preflight
Inspect actual data inside the container
Inspect expected loader contract
Map the gap
Implement adapter/loader
Verify
| Goal | Approach |
|---|---|
| Inspect dataset schema | Use native Python library inside the container |
| Find expected loader | Read dataset classes and config files |
| Validate adapter | Instantiate dataset, fetch sample, assert keys/shapes/dtypes |
| Smoke test training | Same entry point with small batch/few steps |
Use immediately before moving or publishing a checkpoint, and after downloading it, to generate or verify a lightweight SHA-256 file manifest.
Use before running a checkpoint on a robot or inference rig to audit the real sensor-to-action chain using saved model assets, runner code, and live or replay samples.
Use when evaluating checkpoints to maintain a persistent log of what was evaluated, the metrics, qualitative assessment, and verdict.
Use before committing to a full training run to reconcile a base model's baked-in expectations against the training config and the actual dataset, catching semantic mismatches (normalization, feature order, units, encoding) that build and run fine but silently ruin results.
Use when submitting, monitoring, or reviewing HPC training runs to maintain a persistent log of each run's config, status, results, and next steps.
Use when running ML training workflows on Slurm-based HPC clusters, including environment setup, data and image transfer, job submission, monitoring, debugging, and storage cleanup, especially when notes are ad-hoc, usernames differ, or commands may expose secrets.
