config

DeepLearningTaskConfig.

class cneuromax.fitting.deeplearning.config.DeepLearningSubtaskConfig(output_dir='${hydra:runtime.output_dir}', data_dir='${oc.env:CNEUROMAX_PATH}/data/', seed=0, device='cpu', compile=False, save_every_n_train_steps=1, ckpt_path='last')

Bases: FittingSubtaskConfig

Deep Learning subtask config.

Parameters:

• compile (bool, default: False) – Whether to compile the BaseLitModule before training. Requires FittingSubtaskConfig.device to be set to "gpu" & a CUDA 7+ compatible GPU.

• save_every_n_train_steps (int | None, default: 1) – The frequency at which to save training checkpoints.

• ckpt_path (str, default: 'last') – The path to a checkpoint to resume training from.

class cneuromax.fitting.deeplearning.config.DeepLearningTaskConfig(trainer=<class 'types.PartialBuilds_Trainer'>, datamodule=<class 'types.Builds_BaseDataModule'>, litmodule=<class 'types.Builds_BaseLitModule'>, logger=<class 'types.PartialBuilds_WandbLogger'>, config=<class 'types.Builds_DeepLearningSubtaskConfig'>, defaults=<factory>)

Bases: Config

Deep Learning task config.

Parameters:

• defaults (list[Any], default: <factory>) – Hydra defaults.

• trainer (Any, default: <class 'types.PartialBuilds_Trainer'>) – See lightning.pytorch.trainer.trainer.Trainer.

• datamodule (Any, default: <class 'types.Builds_BaseDataModule'>) – See BaseDataModule.

• litmodule (Any, default: <class 'types.Builds_BaseLitModule'>) – See BaseLitModule.

• logger (Any, default: <class 'types.PartialBuilds_WandbLogger'>) – See lightning.pytorch.loggers.wandb.WandbLogger.

• config (Any, default: <class 'types.Builds_DeepLearningSubtaskConfig'>) – See DeepLearningSubtaskConfig.