neuraltrain.models.reve.NtReve¶
- class neuraltrain.models.reve.NtReve(*, kwargs: dict[str, Any] = {}, from_pretrained_name: str | None = None, channel_mapping: dict[str, str] | None = None)[source][source]¶
Config for the braindecode REVE model with channel-mapping support.
Extends
BaseBrainDecodeModelwith REVE-specific logic:Channel remapping – an explicit
channel_mappingdict maps dataset channel names to REVE position-bank names. Channels whose names already appear in the bank (exact match) need no entry.Pretrained loading – bypasses the base-class restriction on
n_timesfor pretrained models, since REVE needs it to size itsfinal_layer.Encoder-only output – when
n_outputsisNone(downstream wrapper handles the head), the model is wrapped to callforward(return_output=True)and return the final transformer layer output, bypassing REVE’sfinal_layer.
- Parameters:
channel_mapping (dict or None) – Explicit mapping from dataset channel names to REVE position-bank names. Useful for EEG systems whose naming convention is absent from the bank (e.g. Neuromag
"EEG 005"or easycap-M10 numeric"2").