Data Preparation

Agent-driven construction of a flat, auditable workspace data view.

1. Boundary

Solver code reads a single workspace-relative root: workspace/dataset/. Raw task data is treated as read-only. Preparation may create metadata and derived splits inside the workspace, while large unchanged media directories are linked rather than copied.

The current protocol does not create Deep/ or Shallow/ subtrees. Legacy prepared inputs are adapted to the flat view by solver/lnr/prep_fs.py.

2. Entry points

ModeBehaviorOutput
scienceflow prepCreates an empty workspace dataset directory, loads the repository skill registry, and runs the data-preparation ScienceAgent.Prepared dataset/, optional split manifest, and a human-readable report.
scienceflow prep --link-onlySkips the agent and symlinks every top-level input entry into the flat dataset view.A fast, non-semantic workspace view.
scienceflow replUses the same flat-symlink fast path when the workspace has no dataset view yet.Dataset access for an interactive agent session.

3. Agent protocol

Inspect Inputread-only task data
List Skillscategory=data_processing
Read Guidancedata_prep plus matching skill
Build Datasetflat files and safe symlinks
Audit Splitmanifest and report

solver/data_prep.py requires the agent to list data-preparation skills, read data_prep, inspect the input root, and write only within the workspace. It uses the standard ScienceAgent with a stable system prompt, bash-write tools, repository skills, and a dedicated memory directory.

4. Validation policy

5. Split artifacts

ArtifactContract
dataset/train*Training-side records and labels after any holdout has been removed.
dataset/validation*Optional local validation records in the schema expected by the task.
dataset/mask_validation_label.*Optional hidden labels when validation features must not expose the target.
dataset/split_manifest.jsonMachine-readable seed, strategy, group key, target columns, counts, distributions, and validity notes.
dataset/split_report.mdShort explanation of the resulting layout, validation policy, and known limitations.

6. Current skill library

The repository currently provides data_prep, image-task guidance, BSON image loading, recommendation time splitting, tabular preprocessing speedups, and self-evaluation guidance under .scienceflow/skills/data_processing/. New dataset knowledge should normally be added as a skill rather than embedded as task-specific branches in the runtime.