Command-line interface¶
RFFM currently has two source-checkout command surfaces. They have different responsibilities and are not installed console scripts.
Bounded pretraining¶
Maturity: verified
python -m rffm.cli.main pretrain
[--config-dir PATH]
[--config-name NAME]
--experiment NAME
--run-id ID
--attempt-id ID
[--rffm-repository-root PATH]
[--gcs-bucket-namespace-root PATH]
[--source-revision SHA]
[--container-image-uri URI]
[--launch-config-uri URI]
[--launch-config-sha256 SHA256]
[--override KEY=VALUE]...
Option |
Required |
Contract |
|---|---|---|
|
no |
Hydra tree; defaults to |
|
no |
Root config name; defaults to |
|
yes |
Readable experiment config-group name |
|
yes |
Logical run identity and artifact path component |
|
yes |
One attempt within the logical run |
|
no |
Root used to resolve repository-relative paths |
|
for |
Directory whose immediate children are mounted bucket names |
provenance options |
no locally |
Source, immutable image, provider launch URI, and launch hash recorded in manifests |
|
no |
Repeatable Hydra dot-list override |
The command writes one sorted JSON result to standard output after success. Configuration, data, validation, or training failures propagate as non-zero process failures. Non-finite training failures also attempt to publish a strict diagnostic and terminal failed manifest without hiding the primary error.
Preprocessing submission¶
Maturity: runtime-isolated
./rffm preprocess DATASET --env dev
This root script submits the selected dataset to the configured Google Cloud
Workflow by invoking gcloud. It prints Workflow, Batch, log, and artifact
identities needed by an operator. The command requires configured Google Cloud
credentials and mutates external state; it is not part of the local
documentation quickstart.
The only current environment is dev. The dataset argument is passed to the
deployed workflow; the local parser does not claim that every registry entry is
canonicalization-ready.