seedmill¶
Config-driven synthetic data generator (local or hosted models).
Usage:
$ seedmill [OPTIONS] COMMAND [ARGS]...
Options:
--help: Show this message and exit.
Commands:
generate: Generate the raw dataset for a task...export: Export split datasets in the task's...stats: Dataset summary: totals, per-group counts,...tasks: List available task configs.init: Scaffold a runnable example task YAML to...
seedmill generate¶
Generate the raw dataset for a task (resumable).
Usage:
$ seedmill generate [OPTIONS]
Options:
-t, --task TEXT: Task YAML path [required]-o, --output TEXT-m, --model TEXT: Override task model--base-url TEXT--backend TEXT: Override backend: ollama | openai_compat (llama.cpp) | openai (hosted)--per-seed INTEGER: Override records per seed-n, --count INTEGER: Override total count (batch mode)--seed TEXT: Only these seed values (repeatable)--limit INTEGER: Only first N seeds (smoke tests)--workers INTEGER: Concurrent seeds (seeded mode); server slots permitting--unconstrained: Disable schema-constrained decoding--help: Show this message and exit.
seedmill export¶
Export split datasets in the task's configured formats.
Usage:
$ seedmill export [OPTIONS]
Options:
-t, --task TEXT: [required]-i, --input TEXT-o, --output-dir TEXT-f, --format TEXT: Formats (repeatable)--seed INTEGER: Split seed override--help: Show this message and exit.
seedmill stats¶
Dataset summary: totals, per-group counts, top enum values.
Usage:
$ seedmill stats [OPTIONS]
Options:
-t, --task TEXT: [required]-i, --input TEXT--help: Show this message and exit.
seedmill tasks¶
List available task configs.
Usage:
$ seedmill tasks [OPTIONS]
Options:
--dir TEXT: [default: config/tasks]--help: Show this message and exit.
seedmill init¶
Scaffold a runnable example task YAML to edit into your own.
Usage:
$ seedmill init [OPTIONS] [NAME]
Arguments:
[NAME]: Name for the scaffolded task [default: example]
Options:
--dir TEXT: [default: config/tasks]--force: Overwrite an existing file--help: Show this message and exit.