batched_inception_time#

torch_uncertainty.models.batched_inception_time(in_channels, num_classes, num_estimators, kernel_size=40, embed_dim=32, num_blocks=6, dropout=0.0, residual=True, repeat_strategy='paper')[source]#

BatchEnsemble of InceptionTime.

Parameters:

  • in_channels (int) – Number of input channels.

  • num_classes (int) – Number of output classes.

  • num_estimators (int) – Number of estimators for BatchEnsemble.

  • kernel_size (int) – Size of the convolutional kernels. Default is 40.

  • embed_dim (int) – Dimension of the embedding. Default is 32.

  • num_blocks (int) – Number of inception blocks. Default is 6.

  • dropout (float) – Dropout rate. Default is 0.0.

  • residual (bool) – Whether to use residual connections. Default is True.

  • repeat_strategy ("legacy"|"paper") –

    The repeat strategy to use during training:

    • ”legacy”: Repeat inputs for each estimator during both training and evaluation.

    • ”paper”(default): Repeat inputs for each estimator only during evaluation.

Returns:

An instance of the InceptionTime model.

Return type:

_BatchedInceptionTime