Skip to content

Quickstart

1. Generate a Sample Config

autonima create-sample-config > config.yaml

The generated file is the canonical sample config used by the docs and repository example.

2. Edit the Config

At minimum, update:

  • search.query
  • screening.abstract.objective
  • screening.abstract.inclusion_criteria
  • screening.fulltext.objective
  • screening.fulltext.inclusion_criteria

If you want to skip a screening stage, set skip_stage: true for that stage instead of leaving required fields empty.

3. Validate Before Running

autonima validate config.yaml

You can also validate with an explicit output folder:

autonima validate config.yaml runs/my_review

4. Run the Pipeline

autonima run config.yaml

Or with an explicit output folder:

autonima run config.yaml runs/my_review

Default Output Folder

If you omit OUTPUT_FOLDER, the CLI derives it from the config filename:

  • Input config: projects/cue_reactivity/default.yaml
  • Default output folder: projects/cue_reactivity/default/

This is separate from the output.directory value stored in the YAML. The CLI will override that runtime path.

5. Run Meta-Analysis

autonima meta expects the folder containing nimads_studyset.json and nimads_annotation.json, which usually means the pipeline outputs/ directory:

autonima meta projects/cue_reactivity/default/outputs

Install the meta extra first:

pip install -e .[meta]

Next Steps