Skip to main content
This page explains the high-level structure of the YAML run config used by Luna Studio.
The SDK reads a single YAML file that contains both data_generation and training settings. You typically run one or both of these steps from the same config:
  • run_data_generation(config_path=...)
  • run_training(config_path=...)

​
Run config structure

Your YAML file controls the overall workflow and includes top-level keys plus nested data_generation and training sections: 
run_steps: ["data_generation", "training"]
pipeline_provider: "local"
metric_name: "custom"

data_generation: ...

training: ...

​
Top-level fields

​
run_steps

run_steps controls which parts of the workflow run. Valid values are:
  • data_generation
  • training
Common patterns:
  • ["data_generation", "training"]: run the full end-to-end workflow
  • ["data_generation"]: generate or label data only
  • ["training"]: skip data generation and train from an existing labelled dataset

​
pipeline_provider

pipeline_provider controls where the workflow runs. This selects the execution backend for your run. Since you are using the SDK, you can fix this to “local”.

​
metric_name

metric_name selects the packaged metric config to use for the run. Supported values include:
  • action_advancement
  • action_completion
  • context_adherence
  • context_relevance
  • prompt_injection
  • sexism
  • tone
  • tool_error_rate
  • tool_selection_quality
  • toxicity
  • custom
Preset metrics provide a starting configuration for common Luna metrics. Use custom when your metric does not match one of the packaged presets. Preset metric example: Using a preset metric

​
Using custom

Set metric_name: "custom" when you want to define your own metric behavior. With custom, you are expected to define the metric explicitly in your config, including:
  • data_generation.metric.name
  • data_generation.metric.description
  • data_generation.metric.type
  • data_generation.metric.input_format
  • data_generation.metric.class_labels or data_generation.metric.llmaj_source_prompt
  • data_generation.source_data.dataset.columns.features
  • training.prompt_template
The packaged custom template starts as a binary setup, uses boolean training output. You should update those values to match your use case. Custom metric example: Trace input / output only

​
Nested config sections

​
data_generation

The data_generation section controls how source data is loaded, how synthetic or labelled examples are produced, and where the resulting dataset is written.

​
training

The training section controls how the model is fine-tuned, evaluated, and where training artifacts are saved.

​
Next steps

For full field-by-field reference, continue to: