Modelpublic

braindecode/CTNet

sourceHugging Facebsd-3-clauseupdated 5mo agoView on Hugging Face

0likes

Model Card

CTNet

CTNet from Zhao, W et al (2024) [ctnet].

Architecture-only repository. Documents the braindecode.models.CTNet class. No pretrained weights are distributed here. Instantiate the model and train it on your own data.

Quick start

bash

pip install braindecode

python

from braindecode.models import CTNet

model = CTNet(
    n_chans=22,
    sfreq=250,
    input_window_seconds=4.0,
    n_outputs=4,
)

The signal-shape arguments above are illustrative defaults — adjust to match your recording.

Documentation

—Full API reference: <https://braindecode.org/stable/generated/braindecode.models.CTNet.html>
—Interactive browser (live instantiation, parameter counts): <https://huggingface.co/spaces/braindecode/model-explorer>
—Source on GitHub: <https://github.com/braindecode/braindecode/blob/master/braindecode/models/ctnet.py#L27>

Architecture

CTNet architecture

Parameters

Parameter	Type	Description
`activation`	nn.Module, default=nn.GELU	Activation function to use in the network.
`num_heads`	int, default=4	Number of attention heads in the Transformer encoder.
`embed_dim`	int or None, default=None	Embedding size (dimensionality) for the Transformer encoder.
`num_layers`	int, default=6	Number of encoder layers in the Transformer.
`n_filters_time`	int, default=20	Number of temporal filters in the first convolutional layer.
`kernel_size`	int, default=64	Kernel size for the temporal convolutional layer.
`depth_multiplier`	int, default=2	Multiplier for the number of depth-wise convolutional filters.
`pool_size_1`	int, default=8	Pooling size for the first average pooling layer.
`pool_size_2`	int, default=8	Pooling size for the second average pooling layer. cnndropprob: float, default=0.3 Dropout probability after convolutional layers.
`att_positional_drop_prob`	float, default=0.1	Dropout probability for the positional encoding in the Transformer.
`final_drop_prob`	float, default=0.5	Dropout probability before the final classification layer.

References

1.Zhao, W., Jiang, X., Zhang, B., Xiao, S., & Weng, S. (2024). CTNet: a convolutional transformer network for EEG-based motor imagery classification. Scientific Reports, 14(1), 20237.
2.Zhao, W., Jiang, X., Zhang, B., Xiao, S., & Weng, S. (2024). CTNet source code: https://github.com/snailpt/CTNet

Citation

Cite the original architecture paper (see References above) and braindecode:

bibtex

@article{aristimunha2025braindecode,
  title   = {Braindecode: a deep learning library for raw electrophysiological data},
  author  = {Aristimunha, Bruno and others},
  journal = {Zenodo},
  year    = {2025},
  doi     = {10.5281/zenodo.17699192},
}

License

BSD-3-Clause for the model code (matching braindecode). Pretraining-derived weights, if you fine-tune from a checkpoint, inherit the licence of that checkpoint and its training corpus.