Add architecture-only model card

README.md

@@ -0,0 +1,246 @@
+---
+license: bsd-3-clause
+library_name: braindecode
+pipeline_tag: feature-extraction
+tags:
+  - eeg
+  - biosignal
+  - pytorch
+  - neuroscience
+  - braindecode
+  - convolutional
+---
+
+# EEGNet
+
+EEGNet model from Lawhern et al (2018) .
+
+> **Architecture-only repository.** This repo documents the
+> `braindecode.models.EEGNet` class. **No pretrained weights are
+> distributed here** — instantiate the model and train it on your own
+> data, or fine-tune from a published foundation-model checkpoint
+> separately.
+
+## Quick start
+
+```bash
+pip install braindecode
+```
+
+```python
+from braindecode.models import EEGNet
+
+model = EEGNet(
+    n_chans=22,
+    sfreq=250,
+    input_window_seconds=4.0,
+    n_outputs=4,
+)
+```
+
+The signal-shape arguments above are example defaults — adjust them
+to match your recording.
+
+## Documentation
+
+- Full API reference (parameters, references, architecture figure):
+  <https://braindecode.org/stable/generated/braindecode.models.EEGNet.html>
+- Interactive browser with live instantiation:
+  <https://huggingface.co/spaces/braindecode/model-explorer>
+- Source on GitHub: <https://github.com/braindecode/braindecode/blob/master/braindecode/models/eegnet.py#L22>
+
+## Architecture description
+
+The block below is the rendered class docstring (parameters,
+references, architecture figure where available).
+
+<div class='bd-doc'><main>
+<p>EEGNet model from Lawhern et al (2018) [Lawhern2018]_.</p>
+<span style="display:inline-block;padding:2px 8px;border-radius:4px;background:#5cb85c;color:white;font-size:11px;font-weight:600;margin-right:4px;">Convolution</span>
+
+
+
+ .. figure:: https://content.cld.iop.org/journals/1741-2552/15/5/056013/revision2/jneaace8cf01_hr.jpg
+     :align: center
+     :alt: EEGNet Architecture
+     :width: 600px
+
+ .. rubric:: Architectural Overview
+
+ EEGNet is a compact convolutional network designed for EEG decoding with a pipeline that mirrors classical EEG processing:
+ - (i) learn temporal frequency-selective filters,
+ - (ii) learn spatial filters for those frequencies, and
+ - (iii) condense features with depthwise-separable convolutions before a lightweight classifier.
+
+ The architecture is deliberately small (temporal convolutional and spatial patterns) [Lawhern2018]_.
+
+ .. rubric:: Macro Components
+
+ - **Temporal convolution**
+   Temporal convolution applied per channel; learns ``F1`` kernels that act as data-driven band-pass filters.
+ - **Depthwise Spatial Filtering.**
+   Depthwise convolution spanning the channel dimension with ``groups = F1``,
+   yielding ``D`` spatial filters for each temporal filter (no cross-filter mixing).
+ - **Norm-Nonlinearity-Pooling (+ dropout).**
+   Batch normalization → ELU → temporal pooling, with dropout.
+ - **Depthwise-Separable Convolution Block.**
+   (a) depthwise temporal conv to refine temporal structure;
+   (b) pointwise 1x1 conv to mix feature maps into ``F2`` combinations.
+ - **Classifier Head.**
+   Lightweight 1x1 conv or dense layer (often with max-norm constraint).
+
+ .. rubric:: Convolutional Details
+
+ - **Temporal.** The initial temporal convs serve as a *learned filter bank*:
+   long 1-D kernels (implemented as 2-D with singleton spatial extent) emphasize oscillatory bands and transients.
+   Because this stage is linear prior to BN/ELU, kernels can be analyzed as FIR filters to reveal each feature's spectrum [Lawhern2018]_.
+
+ - **Spatial.** The depthwise spatial conv spans the full channel axis (kernel height = #electrodes; temporal size = 1).
+   With ``groups = F1``, each temporal filter learns its own set of ``D`` spatial projections—akin to CSP, learned end-to-end and
+   typically regularized with max-norm.
+
+ - **Spectral.** No explicit Fourier/wavelet transform is used. Frequency structure
+   is captured implicitly by the temporal filter bank; later depthwise temporal kernels act as short-time integrators/refiners.
+
+ .. rubric:: Additional Comments
+
+ - **Filter-bank structure:** Parallel temporal kernels (``F1``) emulate classical filter banks; pairing them with frequency-specific spatial filters
+   yields features mappable to rhythms and topographies.
+ - **Depthwise & separable convs:** Parameter-efficient decomposition (depthwise + pointwise) retains power while limiting overfitting
+   [Chollet2017]_ and keeps temporal vs. mixing steps interpretable.
+ - **Regularization:** Batch norm, dropout, pooling, and optional max-norm on spatial kernels aid stability on small EEG datasets.
+ - The v4 means the version 4 at the arxiv paper [Lawhern2018]_.
+
+
+ Parameters
+ ----------
+ final_conv_length : int or "auto", default="auto"
+     Length of the final convolution layer. If "auto", it is set based on n_times.
+ pool_mode : {"mean", "max"}, default="mean"
+     Pooling method to use in pooling layers.
+ F1 : int, default=8
+     Number of temporal filters in the first convolutional layer.
+ D : int, default=2
+     Depth multiplier for the depthwise convolution.
+ F2 : int or None, default=None
+     Number of pointwise filters in the separable convolution. Usually set to ``F1 * D``.
+ depthwise_kernel_length : int, default=16
+     Length of the depthwise convolution kernel in the separable convolution.
+ pool1_kernel_size : int, default=4
+     Kernel size of the first pooling layer.
+ pool2_kernel_size : int, default=8
+     Kernel size of the second pooling layer.
+ kernel_length : int, default=64
+     Length of the temporal convolution kernel.
+ conv_spatial_max_norm : float, default=1
+     Maximum norm constraint for the spatial (depthwise) convolution.
+ activation : nn.Module, default=nn.ELU
+     Non-linear activation function to be used in the layers.
+ batch_norm_momentum : float, default=0.01
+     Momentum for instance normalization in batch norm layers.
+ batch_norm_affine : bool, default=True
+     If True, batch norm has learnable affine parameters.
+ batch_norm_eps : float, default=1e-3
+     Epsilon for numeric stability in batch norm layers.
+ drop_prob : float, default=0.25
+     Dropout probability.
+ final_layer_with_constraint : bool, default=False
+     If ``False``, uses a convolution-based classification layer. If ``True``,
+     apply a flattened linear layer with constraint on the weights norm as the final classification step.
+ norm_rate : float, default=0.25
+     Max-norm constraint value for the linear layer (used if ``final_layer_conv=False``).
+
+ References
+ ----------
+ .. [Lawhern2018] Lawhern, V. J., Solon, A. J., Waytowich, N. R., Gordon, S. M.,
+     Hung, C. P., & Lance, B. J. (2018). EEGNet: a compact convolutional
+     neural network for EEG-based brain–computer interfaces. Journal of
+     neural engineering, 15(5), 056013.
+ .. [Chollet2017] Chollet, F., *Xception: Deep Learning with Depthwise Separable
+     Convolutions*, CVPR, 2017.
+
+ .. rubric:: 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:**
+
+ .. code::
+     from braindecode.models import EEGNet
+
+     # Train your model
+     model = EEGNet(n_chans=22, n_outputs=4, n_times=1000)
+     # ... training code ...
+
+     # Push to the Hub
+     model.push_to_hub(
+         repo_id="username/my-eegnet-model",
+         commit_message="Initial model upload",
+     )
+
+ **Loading a model from the Hub:**
+
+ .. code::
+     from braindecode.models import EEGNet
+
+     # Load pretrained model
+     model = EEGNet.from_pretrained("username/my-eegnet-model")
+
+     # Load with a different number of outputs (head is rebuilt automatically)
+     model = EEGNet.from_pretrained("username/my-eegnet-model", n_outputs=4)
+
+ **Extracting features and replacing the head:**
+
+ .. code::
+     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:**
+
+ .. code::
+     import json
+
+     config = model.get_config()            # all __init__ params
+     with open("config.json", "w") as f:
+         json.dump(config, f)
+
+     model2 = EEGNet.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 :ref:`load-pretrained-models` for a complete tutorial.</main>
+</div>
+
+## Citation
+
+Please cite both the original paper for this architecture (see the
+*References* section 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.