braindecode.models.EEGModuleMixin#

class braindecode.models.EEGModuleMixin(n_outputs=None, n_chans=None, chs_info=None, n_times=None, input_window_seconds=None, sfreq=None, channel_strategy='native', channel_strategy_kwargs=None)[source]#

Mixin class for all EEG models in braindecode.

This class integrates with Hugging Face Hub when the huggingface_hub package is installed, enabling models to be pushed to and loaded from the Hub using push_to_hub() and from_pretrained() methods.

Parameters:

n_outputs (Optional[int]) – Number of outputs of the model. This is the number of classes in the case of classification.
n_chans (Optional[int]) – Number of EEG channels.
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.
n_times (Optional[int]) – Number of time samples of the input window.
input_window_seconds (Optional[float]) – Length of the input window in seconds.
sfreq (Optional[float]) – Sampling frequency of the EEG recordings.
channel_strategy (str) – How any montage reaches the backbone (pretrained models only; see Channel strategies: any montage in). "native" keeps the model as it is. "exact", "zero", "nearest", "idw", "spline", "field", "source", "region", "wiener" (call model.channel_layer.fit first) or "latent" map the montage of chs_info (or of the chs_info given to forward) onto the backbone’s channels with a ChannelLayer. Saved in the config. model(x) and model.forward(x) both apply the layer.
channel_strategy_kwargs (Optional[dict]) – Options of the strategy (e.g. {"reg": 1e-2} for "spline").

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.

Methods

classmethod from_config(config)[source]#

Create a model instance from a configuration dict.

This is the inverse of get_config(). Weights are not loaded – use from_pretrained() for that.

Parameters:: config (dict) – Configuration dict as returned by get_config().
Returns:: A new model instance.
Return type:: EEGModuleMixin

Examples

>>> import json
>>> from braindecode.models import EEGNet
>>> model = EEGNet(n_chans=22, n_times=1000, n_outputs=4, F1=16)
>>> config = model.get_config()
>>> # Reconstruct (without weights)
>>> model2 = EEGNet.from_config(config)
>>> model2.F1
16
>>> # Or from a JSON file
>>> with open("config.json") as f:
...     config = json.load(f)
>>> model3 = EEGNet.from_config(config)

Added in version 1.4.

classmethod from_pretrained(*args, **kwargs)[source]#

Download a model from the Huggingface Hub and instantiate it.

Parameters:

pretrained_model_name_or_path (str, Path) –
- Either the model_id (string) of a model hosted on the Hub, e.g. bigscience/bloom.
- Or a path to a directory containing model weights saved using
  [~transformers.PreTrainedModel.save_pretrained], e.g., ../path/to/my_model_directory/.
revision (str, optional) – Revision of the model on the Hub. Can be a branch name, a git tag or any commit id. Defaults to the latest commit on main branch.
force_download (bool, optional, defaults to False) – Whether to force (re-)downloading the model weights and configuration files from the Hub, overriding the existing cache.
token (str or bool, optional) – The token to use as HTTP bearer authorization for remote files. By default, it will use the token cached when running hf auth login.
cache_dir (str, Path, optional) – Path to the folder where cached files are stored.
local_files_only (bool, optional, defaults to False) – If True, avoid downloading the file and return the path to the local cached file if it exists.
model_kwargs (dict, optional) – Additional kwargs to pass to the model during initialization.
*args – The description is missing.
**kwargs – The description is missing.

get_config()[source]#

Return a JSON-serializable dict of all __init__ parameters.

The returned dictionary can be saved to a JSON file and later used with from_config() to reconstruct the model (without weights). It is also used internally by push_to_hub() to persist the full model configuration.

Returns:: All __init__ parameters, JSON-serializable. type[nn.Module] parameters (e.g. activation) are encoded as importable dotted-path strings.
Return type:: dict

Examples

>>> import json
>>> from braindecode.models import EEGNet
>>> model = EEGNet(n_chans=22, n_times=1000, n_outputs=4, F1=16)
>>> config = model.get_config()
>>> config["F1"]
16
>>> # Save to disk
>>> with open("config.json", "w") as f:
...     json.dump(config, f)

Added in version 1.4.

get_output_shape()[source]#

Returns shape of neural network output for batch size equal 1.

Returns:: output_shape – shape of the network output for batch_size==1 (1, …)
Return type:: tuple[int, ...]

get_torchinfo_statistics(col_names=('input_size', 'output_size', 'num_params', 'kernel_size'), row_settings=('var_names', 'depth'))[source]#

Generate table describing the model using torchinfo.summary.

Parameters:

col_names (Optional[Iterable[str]]) – Specify which columns to show in the output, see torchinfo for details, by default (“input_size”, “output_size”, “num_params”, “kernel_size”)
row_settings (Optional[Iterable[str]]) – Specify which features to show in a row, see torchinfo for details, by default (“var_names”, “depth”)

Returns:

ModelStatistics generated by torchinfo.summary.

Return type:

ModelStatistics

load_state_dict(state_dict, *args, **kwargs)[source]#

reset_head(n_outputs)[source]#

Replace the classification head for a new number of outputs.

This is called automatically by from_pretrained() when the user passes an n_outputs that differs from the saved config. Override in subclasses that need a model-specific head structure. Implementations keep changed constructor arguments in sync with self._update_init_kwargs, so that a saved model can be loaded back. Implementations requiring positive outputs can also use self._set_n_outputs to validate and record the new value.

Parameters:: n_outputs (int) – New number of output classes.

Examples

>>> from braindecode.models import BENDR
>>> model = BENDR(n_chans=22, n_times=1000, n_outputs=4)
>>> model.reset_head(10)
>>> model.n_outputs
10

Added in version 1.4.

to_dense_prediction_model(axis=(2, 3))[source]#

Transform a sequential model with strides to a model that outputs.

dense predictions by removing the strides and instead inserting dilations. Modifies model in-place.

Parameters:: axis (tuple[int, ...] | int) – Axis to transform (in terms of intermediate output axes) can either be 2, 3, or (2,3).
Return type:: None