CondVAEModel.evaluate

CondVAEModel.evaluate(datamodule: DataModule, unnormalize: bool = False) → Tuple[DataFrame, DataFrame, DataFrame, DataFrame][source]

Evaluate the model on the validation data. This method is similar to CondAEModel.validate(), but it uses DataObject.get_loss_evaluation() instead of DataObject.get_objective() to compute the losses. It computes the losses per sample rather than summing/averaging them over the entire validation set.

Parameters:

  • datamodule (DataModule) – The data module object that provides validation data.

  • unnormalize (bool, optional, default=False) – If True, the losses are computed in the unnormalized space.

Returns:

Tuple[pd.DataFrame, pd.DataFrame, pd.DataFrame, pd.DataFrame] – A tuple containing the evaluation losses for the input and output features, as well as the values predicted for input and output features.

Notes

This method, for now, only works on CPU, as the unnormalization is done on the CPU.