菜单
TokSearch compute backends — serial, multiprocessing, Ray, and Spark; when to use each and configuration options
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
TokSearch xarray dataset integration — fetch_dataset and align for multi-signal time-aligned analysis
MdsSignal for fetching MDSplus tree data — constructor args, location formats, TDI expressions, dims, and DIII-D efit01 signals via FDP/Pelican
Core TokSearch Pipeline workflow — creating pipelines, fetch/map/where/keep, record data access, and iterating results
| name | toksearch-backends |
| description | TokSearch compute backends — serial, multiprocessing, Ray, and Spark; when to use each and configuration options |
| user-invocable | false |
| license | Apache-2.0 |
| compatibility | Claude Code |
| metadata | {"author":"GA-FDP","version":"2.0","url":"https://ga-fdp.github.io/toksearch/"} |
The canonical backend documentation lives in the toksearch package docstring.
Access it with: help(toksearch) or python -c "help(toksearch)"
Covers: serial, multiprocessing, Ray, and Spark backends; decision guide; RecordSet iteration; and pandas DataFrame conversion.
