braindecode.visualization.amplitude_gradients_per_trial#
- braindecode.visualization.amplitude_gradients_per_trial(model, dataset, batch_size)[source]#
Concatenated
amplitude_gradients()over every trial in a dataset.- Parameters:
model (torch.nn.Module)
dataset (torch.utils.data.Dataset) – Yields
(x, ...)tuples; only the first element is used.batch_size (int)
- Returns:
Per-trial amplitude gradients for each output unit, in dataset order.
- Return type:
numpy.ndarray of shape
(n_outputs, n_trials, n_chans, n_freqs)
