mlvbench.infrastructure¶
Infrastructure utilities.
resolve_layers(all_layers, layers_spec)
¶
Resolve a layer specification string to a list of layer names.
Intended for experiment scripts that expose a --layers command line argument.
The resolved list can be passed directly to
ProbeTrainer.fit.
Parameters:
| Name | Type | Description | Default |
|---|---|---|---|
all_layers
|
list[str]
|
The full list of layer names from the model
( |
required |
layers_spec
|
str | None
|
A specification of a list of layers, provided as a comma-separated string. Each value may be a non-negative index (e.g. 0), a negative index (e.g. -1), or a layer name (e.g. block.12). If None, all layers are returned. |
required |
Returns:
| Type | Description |
|---|---|
list[str]
|
A deduplicated list of resolved layer names. |
save_metadata(output_path, parameters)
¶
Save the metadata for an experiment run.
save_model_info(output_path, model)
¶
Save model info for an experiment run.