--- library_name: transformers tags: - eeg - neuroscience - foundation-model - pytorch license: other license_name: reve-responsible-use-license-v1.0 license_link: LICENSE model-index: - name: Reve-base results: - task: type: feature-extraction dataset: name: TUAB type: TUAB metrics: - type: Accuracy value: 0.8315 name: Accuracy - task: type: feature-extraction dataset: name: TUEV type: TUEV metrics: - type: Accuracy value: 0.6759 name: Accuracy - task: type: feature-extraction dataset: name: PhysionetMI type: PhysionetMI metrics: - type: Accuracy value: 0.648 name: Accuracy - task: type: feature-extraction dataset: name: BCICIV2a type: BCICIV2a metrics: - type: Accuracy value: 0.6396 name: Accuracy - task: type: feature-extraction dataset: name: FACED type: FACED metrics: - type: Accuracy value: 0.5646 name: Accuracy - task: type: feature-extraction dataset: name: ISRUC type: ISRUC metrics: - type: Accuracy value: 0.7819 name: Accuracy - task: type: feature-extraction dataset: name: Mumtaz type: Mumtaz metrics: - type: Accuracy value: 0.9644 name: Accuracy - task: type: feature-extraction dataset: name: MentalArithmetic type: MentalArithmetic metrics: - type: Accuracy value: 0.766 name: Accuracy - task: type: feature-extraction dataset: name: BCI2020-3 type: BCI2020-3 metrics: - type: Accuracy value: 0.5635 name: Accuracy - task: type: feature-extraction dataset: name: TUAB-LP type: TUAB-LP metrics: - type: Accuracy value: 0.81 name: Accuracy - task: type: feature-extraction dataset: name: TUEV-LP type: TUEV-LP metrics: - type: Accuracy value: 0.592 name: Accuracy - task: type: feature-extraction dataset: name: PhysionetMI-LP type: PhysionetMI-LP metrics: - type: Accuracy value: 0.537 name: Accuracy - task: type: feature-extraction dataset: name: BCICIV2a-LP type: BCICIV2a-LP metrics: - type: Accuracy value: 0.517 name: Accuracy - task: type: feature-extraction dataset: name: ISRUC-LP type: ISRUC-LP metrics: - type: Accuracy value: 0.697 name: Accuracy - task: type: feature-extraction dataset: name: Mumtaz-LP type: Mumtaz-LP metrics: - type: Accuracy value: 0.962 name: Accuracy - task: type: feature-extraction dataset: name: MentalArithmetic-LP type: MentalArithmetic-LP metrics: - type: Accuracy value: 0.74 name: Accuracy - task: type: feature-extraction dataset: name: BCII2020-3-LP type: BCII2020-3-LP metrics: - type: Accuracy value: 0.39 name: Accuracy --- # Model Card for REVE-base REVE ([project page here](https://brain-bzh.github.io/reve/)) is a transformer-based foundation model for EEG signal processing. It was trained on 60k hours of EEG data from various sources and is designed to be adaptable to any electrode configuration and a wide range of EEG-based tasks. ## Model Details ### Architecture REVE (Representation for EEG with Versatile Embeddings), a pretrained encoder explicitly designed to generalize across diverse EEG signals. REVE introduces a novel 4D positional encoding scheme that enables it to process signals of arbitrary length and electrode arrangement. Using a masked autoencoding objective, we pretrain REVE on over 60,000 hours of EEG data from 92 datasets spanning 25,000 subjects. **Developed by** the [BRAIN team](https://www.imt-atlantique.fr/en/research-innovation/teams/brain) and [UdeM](https://www.umontreal.ca/en/) **Funded by:** This research was supported by the French National Research Agency (ANR) through its AI@IMT program and grant ANR-24-CE23-7365, as well as by a grant from the Brittany region. Further support was provided by a Discovery Grant from the Natural Sciences and Engineering Research Council of Canada (NSERC), by funding from the Canada Research Chairs program and the Fonds de recherche du Québec – Nature et technologies (FRQ-NT). This work was granted access to the HPC resources of IDRIS under the allocation 2024-AD011015237R1 made by GENCI, as well as HPC provided by Digital Alliance Canada. ### Model Sources - **Repository:** [github](https://brain-bzh.github.io/reve/) - **Paper :** [arxiv](https://arxiv.org/abs/2510.21585) ## Uses Example script to extract embeddings with REVE, using our position bank: ```python from transformers import AutoModel pos_bank = AutoModel.from_pretrained("brain-bzh/reve-positions", trust_remote_code=True) model = AutoModel.from_pretrained("brain-bzh/reve-base", trust_remote_code=True) eeg_data = ... # EEG data as a torch Tensor (batch_size, channels, time_points), must be sampled at 200 Hz electrode_names = [...] # List of electrode names corresponding to the channels in eeg_data positions = pos_bank(electrode_names) # Get positions (channels, 3) # Expand the positions vector to match the batch size positions = positions.expand(eeg_data.size(0), -1, -1) # (batch_size, channels, 3) output = model(eeg_data, positions) ``` ## License and Responsible Use REVE is available for research, commercial, educational, and personal use under the **REVE Responsible Use License v1.0**. By downloading, using, modifying, redistributing, or deploying REVE or a derivative model, you agree to comply with the terms of the [`LICENSE`](./LICENSE). In particular, the License prohibits privacy intrusion and re-identification, non-consensual surveillance or profiling, discriminatory or harmful uses, and other uses that violate applicable law. Redistribution and publication of fine-tuned or otherwise modified REVE models are permitted, provided that the License, attribution, and provenance requirements are preserved.