DVC Migration

Crab Workflow intentionally keeps the familiar DVC shape: stages, deps, outs, params, metrics, plots, experiments, queues, and repro-style execution. The main difference is the storage model. Crab uses your Crab remote and object store rather than a separate DVC remote.

For the automated migration command, see the full Migrating from DVC guide.

Command Mapping

Migration Checklist

1. Commit or stash current DVC work.

2. Initialize Crab and enable Workflow:

   crab init crab://my-bucket/ml-project
   crab config set workflow.enabled true

3. Convert dvc.yaml:

   crab migrate from-dvc

4. Validate:

   crab run --validate
   crab workflow dag

5. Run once to create crab.lock:

   crab run

6. Push git state and stage cache:

   git add crab.yaml crab.lock params.yaml
   git commit -m "migrate workflow to crab"
   crab push
   crab workflow push-cache --all

7. Remove DVC files after the Crab workflow is verified.

YAML Compatibility

Most DVC fields map directly:

Behavior Differences

• Crab Workflow must be enabled with crab config set workflow.enabled true.
• Stage cache is shared through the Crab remote with crab run --cache-push or crab workflow push-cache --all.
• Large files should be tracked with Crab's file layer, not a DVC remote.
• Local experiment metadata lives under .crab/workflow/exp/.
• Run journals live under .crab/workflow/runs/.
• Split workflow files can use workflow.discover recursive and workflow.lockfile split.

Hydra

Enable Hydra-style composition before migrating Hydra-heavy experiments:

crab config set hydra.enabled true
crab config set hydra.config_dir conf
crab config set hydra.config_name config.yaml

Then run experiments with group and scalar overrides:

crab exp run -S train/model=efficientnet -S train.optimizer.lr=0.0005

See Hydra Workflows for the recommended layout and precedence model.

CI Migration

Replace DVC cache commands with Crab cache commands:

crab run --validate
crab run --pull --allow-missing --cache-push --jsonl
crab workflow push-cache --all --json

For strict read-only verification:

crab run --cache-only --json

Final Cutover

Only remove DVC files after a clean Crab run, metrics check, and remote cache push:

crab run
crab metrics show
crab plots show
crab workflow push-cache --all

Then remove DVC-specific files and update CI scripts. Keep the DVC branch around until the first Crab-backed release is reproducible from a fresh clone.