fugue
Operate Fugue workflows, FugueSQL, backend selection, and notebook magics for pandas and distributed dataframe execution.
- License 许可
- Apache-2.0
- 最近更新
- 2026-09-03
- Stars
- ★ 351
- Forks
- 27
01 前言 Front Matter
- name
- fugue
- license
- Apache 2.0
- metadata
- { "disco-role": "operating" }
- description
- Operate Fugue workflows, FugueSQL, backend selection, and notebook magics for pandas and distributed dataframe execution.
- disable-model-invocation
- true
02 资源文件 5
- 参考文档references/overview.md3.5 KB
- 参考文档references/repo-provenance.md1.0 KB
- 参考文档references/repo-routing-metadata.json398 B
- 参考文档references/troubleshooting.md2.5 KB
- 脚本scripts/inspect_install.py5.4 KB
03 SKILL.md 正文
Fugue
Use this skill when a request is about Fugue's unified dataframe and SQL workflow API. Fugue wraps pandas-style code and runs it on Native, Spark, Dask, Ray, DuckDB, Ibis, and Polars backends.
Start here
- Read
references/overview.mdfor install choices and the top-level API map. - Read
references/troubleshooting.mdfor generic import, extra, and runtime failures. - Read
references/repo-provenance.mdwhen you need to judge staleness. - Run
scripts/inspect_install.py --helporscripts/inspect_install.pyto confirm the installed package. - Then choose the narrowest sub-skill:
sub-skills/workflow/forFugueWorkflow,transform,process,output, partitioning, checkpoints,module, and dataframe helpers.sub-skills/sql/forfugue_sql,fsql,fugue_sql_flow,raw_sql,YIELD,PRINT,TRANSFORM USING,OUTPUT USING, and SQL templating.sub-skills/backends/for execution-engine selection, registration, plugin imports, and backend extras.sub-skills/notebook/for%load_ext fugue_notebook,%%fsql,NotebookSetup, and Jupyter display behavior.
Quick install guide
Install the smallest extra that matches the task:
pip install fuguefor core dataframe and workflow helpers.pip install "fugue[sql]"for FugueSQL.pip install "fugue[duckdb]","fugue[spark]","fugue[dask]","fugue[ray]","fugue[ibis]","fugue[polars]", or"fugue[notebook]"for a single backend or notebook support.pip install "fugue[all]"when you want the broad runtime set.
Minimal import check
python -c "import fugue, fugue.api as fa; print(fugue.__version__)"
If the request mentions a backend, notebook, or FugueSQL syntax error, open the matching sub-skill before writing code.
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).