ise.models.predictors package

Submodules

ise.models.predictors.deep_ensemble module

class ise.models.predictors.deep_ensemble.DeepEnsemble(ensemble_members=None, input_size=83, output_size=1, num_ensemble_members=3, output_sequence_length=86, latent_dim=1)

Bases: Module

fit(X, y, X_val=None, y_val=None, save_checkpoints=True, checkpoint_path='checkpoint_ensemble', early_stopping=False, epochs=100, batch_size=128, sequence_length=5, patience=10, verbose=True)

Trains the ensemble with optional early stopping.

Args: - X, y: Training data. - early_stopping (bool): Use early stopping. Defaults to False.

forward(x)

Performs a forward pass through the ensemble.

Args: - x: Input data.

Returns: - mean_prediction: Mean prediction across ensemble members. - epistemic_uncertainty: Standard deviation across ensemble predictions.

classmethod load(model_path)

predict(x)

Makes predictions using the ensemble.

Args: - x: Input data.

Returns: - Tuple[Tensor, Tensor]: Mean predictions and uncertainty estimates.

save(model_path)

ise.models.predictors.lstm module

class ise.models.predictors.lstm.LSTM(lstm_num_layers, lstm_hidden_size, input_size=83, output_size=1, criterion=MSELoss(), output_sequence_length=86, optimizer=<class 'torch.optim.adam.Adam'>)

Bases: Module

fit(X, y, epochs=100, sequence_length=5, batch_size=64, criterion=None, X_val=None, y_val=None, save_checkpoints=True, checkpoint_path='checkpoint.pt', early_stopping=False, patience=10, verbose=True, dataclass=<class 'ise.data.dataclasses.EmulatorDataset'>)

forward(x)

Define the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

predict(X, sequence_length=5, batch_size=64, dataclass=<class 'ise.data.dataclasses.EmulatorDataset'>)

Module contents