braindecode.models.ShallowFBCSPNet#

class braindecode.models.ShallowFBCSPNet(n_chans=None, n_outputs=None, n_times=None, n_filters_time=40, filter_time_length=25, n_filters_spat=40, pool_time_length=75, pool_time_stride=15, final_conv_length='auto', conv_nonlin=<class 'braindecode.modules.activation.Square'>, pool_mode='mean', activation_pool_nonlin=<class 'braindecode.modules.activation.SafeLog'>, split_first_layer=True, batch_norm=True, batch_norm_alpha=0.1, drop_prob=0.5, chs_info=None, input_window_seconds=None, sfreq=None)[source]#

Shallow ConvNet model from Schirrmeister et al (2017) [Schirrmeister2017].

Convolution

Model described in [Schirrmeister2017].

Parameters:

n_chans (int) – Number of EEG channels.
n_outputs (int) – Number of outputs of the model. This is the number of classes in the case of classification.
n_times (int) – Number of time samples of the input window.
n_filters_time (int) – Number of temporal filters.
filter_time_length (int) – Length of the temporal filter.
n_filters_spat (int) – Number of spatial filters.
pool_time_length (int) – Length of temporal pooling filter.
pool_time_stride (int) – Length of stride between temporal pooling filters.
final_conv_length (int | str) – Length of the final convolution layer. If set to “auto”, length of the input signal must be specified.
conv_nonlin (type[Module] | Callable) – Non-linear module class to be used after convolution layers. For backward compatibility, callables are also accepted and wrapped with Expression.
pool_mode (str) – Method to use on pooling layers. “max” or “mean”.
activation_pool_nonlin (type[Module]) – Non-linear module class to be used after pooling layers.
split_first_layer (bool) – Split first layer into temporal and spatial layers (True) or just use temporal (False). There would be no non-linearity between the split layers.
batch_norm (bool) – Whether to use batch normalisation.
batch_norm_alpha (float) – Momentum for BatchNorm2d.
drop_prob (float) – Dropout probability.
chs_info (list of dict) – Information about each individual EEG channel. This should be filled with info["chs"]. Refer to mne.Info for more details.
input_window_seconds (float) – Length of the input window in seconds.
sfreq (float) – Sampling frequency of the EEG recordings.

Raises:

ValueError – If some input signal-related parameters are not specified: and can not be inferred.

Notes

If some input signal-related parameters are not specified, there will be an attempt to infer them from the other parameters.

References

[Schirrmeister2017] (1,2)

Schirrmeister, R. T., Springenberg, J. T., Fiederer, L. D. J., Glasstetter, M., Eggensperger, K., Tangermann, M., Hutter, F. & Ball, T. (2017). Deep learning with convolutional neural networks for EEG decoding and visualization. Human Brain Mapping , Aug. 2017. Online: http://dx.doi.org/10.1002/hbm.23730

Hugging Face Hub integration

When the optional huggingface_hub package is installed, all models automatically gain the ability to be pushed to and loaded from the Hugging Face Hub. Install with:

pip install braindecode[hub]

Pushing a model to the Hub:

from braindecode.models import ShallowFBCSPNet

# Train your model
model = ShallowFBCSPNet(n_chans=22, n_outputs=4, n_times=1000)
# ... training code ...

# Push to the Hub
model.push_to_hub(
    repo_id="username/my-shallowfbcspnet-model",
    commit_message="Initial model upload",
)

Loading a model from the Hub:

from braindecode.models import ShallowFBCSPNet

# Load pretrained model
model = ShallowFBCSPNet.from_pretrained("username/my-shallowfbcspnet-model")

# Load with a different number of outputs (head is rebuilt automatically)
model = ShallowFBCSPNet.from_pretrained("username/my-shallowfbcspnet-model", n_outputs=4)

Extracting features and replacing the head:

import torch

x = torch.randn(1, model.n_chans, model.n_times)
# Extract encoder features (consistent dict across all models)
out = model(x, return_features=True)
features = out["features"]

# Replace the classification head
model.reset_head(n_outputs=10)

Saving and restoring full configuration:

import json

config = model.get_config()            # all __init__ params
with open("config.json", "w") as f:
    json.dump(config, f)

model2 = ShallowFBCSPNet.from_config(config)    # reconstruct (no weights)

All model parameters (both EEG-specific and model-specific such as dropout rates, activation functions, number of filters) are automatically saved to the Hub and restored when loading.

See Loading and Adapting Pretrained Foundation Models for a complete tutorial.