ArjunViT

class torcheeg.models.ArjunViT(num_electrodes: int = 32, chunk_size: int = 128, t_patch_size: int = 32, hid_channels: int = 32, depth: int = 3, heads: int = 4, head_channels: int = 64, mlp_channels: int = 64, num_classes: int = 2, embed_dropout: float = 0.0, dropout: float = 0.0, pool_func: str = 'cls')

Arjun et al. employ a variation of the Transformer, the Vision Transformer to process EEG signals for emotion recognition. For more details, please refer to the following information.

It is worth noting that this model is not designed for EEG analysis, but shows good performance and can serve as a good research start.

• Paper: Arjun A, Rajpoot A S, Panicker M R. Introducing attention mechanism for eeg signals: Emotion recognition with vision transformers[C]//2021 43rd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC). IEEE, 2021: 5723-5726.

• URL: https://ieeexplore.ieee.org/abstract/document/9629837

Below is a recommended suite for use in emotion recognition tasks:

from torcheeg.datasets import DEAPDataset
from torcheeg import transforms
from torcheeg.models import ArjunViT
from torch.utils.data import DataLoader

dataset = DEAPDataset(root_path='./data_preprocessed_python',
                      offline_transform=transforms.Compose([
                          transforms.MeanStdNormalize(),
                          transforms.To2d()
                      ]),
                      online_transform=transforms.Compose([
                          transforms.ToTensor(),
                      ]),
                      label_transform=transforms.Compose([
                          transforms.Select('valence'),
                          transforms.Binary(5.0),
                      ]))

model = ArjunViT(chunk_size=128,
                 t_patch_size=50,
                 num_electrodes=32,
                 num_classes=2)

x, y = next(iter(DataLoader(dataset, batch_size=64)))
model(x)

Parameters:

num_electrodes (int) – The number of electrodes. (default: 32) chunk_size (int): Number of data points included in each EEG chunk. (default: 128) t_patch_size (int): The size of each input patch at the temporal (chunk size) dimension. (default: 32) patch_size (tuple): The size (resolution) of each input patch. (default: (3, 3)) hid_channels (int): The feature dimension of embeded patch. (default: 32) depth (int): The number of attention layers for each transformer block. (default: 3) heads (int): The number of attention heads for each attention layer. (default: 4) head_channels (int): The dimension of each attention head for each attention layer. (default: 8) mlp_channels (int): The number of hidden nodes in the fully connected layer of each transformer block. (default: 64) num_classes (int): The number of classes to predict. (default: 2) embed_dropout (float): Probability of an element to be zeroed in the dropout layers of the embedding layers. (default: 0.0) dropout (float): Probability of an element to be zeroed in the dropout layers of the transformer layers. (default: 0.0) pool_func (str): The pool function before the classifier, optionally including cls and mean, where cls represents selecting classification-related token and mean represents the average pooling. (default: cls)

forward(x: Tensor) → Tensor

Parameters:

x (torch.Tensor) – EEG signal representation, the ideal input shape is [n, 32, 128]. Here, n corresponds to the batch size, 32 corresponds to num_electrodes, and chunk_size corresponds to chunk_size.

Returns:

the predicted probability that the samples belong to the classes.

Return type:

torch.Tensor[number of sample, number of classes]