Generate synthetic training data from a YAML configuration file.
(
config_file: str | None,
output_dir: str | None,
export_format: str | None,
task_definition: str | None,
generator_provider: str | None,
generator_model: str | None,
)
| 62 | @click.option("--generator_provider", type=str, help="LLM provider for generator (e.g., openai, ollama)") |
| 63 | @click.option("--generator_model", type=str, help="LLM model name for generator") |
| 64 | def generate( |
| 65 | config_file: str | None, |
| 66 | output_dir: str | None, |
| 67 | export_format: str | None, |
| 68 | task_definition: str | None, |
| 69 | generator_provider: str | None, |
| 70 | generator_model: str | None, |
| 71 | ) -> None: |
| 72 | """Generate synthetic training data from a YAML configuration file.""" |
| 73 | # process configuration from a YAML file. |
| 74 | config = SDGSConfig.from_yaml(config_file) |
| 75 | |
| 76 | # process additional options |
| 77 | config.update(transform_arguments_to_configurations( |
| 78 | output_dir, |
| 79 | export_format, |
| 80 | task_definition, |
| 81 | generator_provider, |
| 82 | generator_model, |
| 83 | )) |
| 84 | |
| 85 | # run pipeline to get data |
| 86 | pipeline = Pipeline(config) |
| 87 | pipeline.run() |
| 88 | |
| 89 | |
| 90 | @cli.command() |
no test coverage detected