replicate_pipeline_from_config
- replicate_pipeline_from_config(config, directory, replicates, move_to_cpu=False, save_replicates=True, save_training=False, keep_seed=False, **kwargs)[source]
Run the same pipeline several times from a configuration dictionary.
- Parameters:
config (
Mapping
[str
,Any
]) – The configuration dictionary for the experiment.replicates (
int
) – The number of replicates to runmove_to_cpu (
bool
) – Should the models be moved back to the CPU? Only relevant if training on GPU.save_replicates (
bool
) – Should the artifacts of the replicates be saved?kwargs – Keyword arguments to be passed through to
pipeline_from_config()
.save_training (
bool
) – whether to save the training factory.keep_seed (
bool
) – whether to keep the random seed in a configuration
- Return type: