MolecularDiffusion.cli.eval_predict¶

Eval-Predict command for MolCraft CLI.

Adapted from scripts/eval_predict.py for package-level execution.

Attributes¶

Functions¶

`engine_wrapper`(task_module, data_module, trainer_module)	Run evaluation with Engine.
`eval_predict_main`(cfg)	Entry point for CLI eval-predict command.
`is_rank_zero`()	Check if current process is rank zero.
`load_checkpoint_weights`(task, chkpt_path)	Load weights from checkpoint with support for Engine and Lightning formats.
`log_hyperparameters`(object_dict)	Log hyperparameters for debugging.
`predict`(→ Tuple[Dict[str, Any], Dict[str, Any]])	Evaluate predictions on validation/test sets.

Module Contents¶

MolecularDiffusion.cli.eval_predict.engine_wrapper(task_module, data_module, trainer_module)¶: Run evaluation with Engine.

MolecularDiffusion.cli.eval_predict.eval_predict_main(cfg: omegaconf.DictConfig)¶: Entry point for CLI eval-predict command.

MolecularDiffusion.cli.eval_predict.is_rank_zero()¶: Check if current process is rank zero.

MolecularDiffusion.cli.eval_predict.load_checkpoint_weights(task, chkpt_path)¶: Load weights from checkpoint with support for Engine and Lightning formats.

MolecularDiffusion.cli.eval_predict.log_hyperparameters(object_dict: dict)¶: Log hyperparameters for debugging.

MolecularDiffusion.cli.eval_predict.predict(cfg: omegaconf.DictConfig) → Tuple[Dict[str, Any], Dict[str, Any]]¶: Evaluate predictions on validation/test sets.

MolecularDiffusion.cli.eval_predict.log¶