braindecode 1.3.0.dev176728557__tar.gz → 1.3.0.dev179592623__tar.gz

{braindecode-1.3.0.dev176728557/braindecode.egg-info → braindecode-1.3.0.dev179592623}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: braindecode
-Version: 1.3.0.dev176728557
+Version: 1.3.0.dev179592623
 Summary: Deep learning software to decode EEG, ECG or MEG signals
 Author-email: Robin Tibor Schirrmeister <robintibor@gmail.com>
 Maintainer-email: Alexandre Gramfort <agramfort@meta.com>, Bruno Aristimunha Pinto <b.aristimunha@gmail.com>, Robin Tibor Schirrmeister <robintibor@gmail.com>

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/braindecode/models/atcnet.py

@@ -370,7 +370,7 @@ class ATCNet(EEGModuleMixin, nn.Module):
                 nn.Sequential(
                     *[
                         _TCNResidualBlock(
-                            in_channels=self.F2,
+                            in_channels=self.F2 if i == 0 else self.tcn_n_filters,
                             kernel_size=self.tcn_kernel_size,
                             n_filters=self.tcn_n_filters,
                             dropout=self.tcn_dropout,
@@ -388,7 +388,7 @@ class ATCNet(EEGModuleMixin, nn.Module):
             self.final_layer = nn.ModuleList(
                 [
                     MaxNormLinear(
-                        in_features=self.F2 * self.n_windows,
+                        in_features=self.tcn_n_filters * self.n_windows,
                         out_features=self.n_outputs,
                         max_norm_val=self.max_norm_const,
                     )
@@ -398,7 +398,7 @@ class ATCNet(EEGModuleMixin, nn.Module):
             self.final_layer = nn.ModuleList(
                 [
                     MaxNormLinear(
-                        in_features=self.F2,
+                        in_features=self.tcn_n_filters,
                         out_features=self.n_outputs,
                         max_norm_val=self.max_norm_const,
                     )
@@ -695,8 +695,8 @@ class _TCNResidualBlock(nn.Module):
         # Reshape the input for the residual connection when necessary
         if in_channels != n_filters:
             self.reshaping_conv = nn.Conv1d(
-                in_channels=in_channels,
-                out_channels=n_filters,
+                in_channels=in_channels,  # Specify input channels
+                out_channels=n_filters,  # Specify output channels
                 kernel_size=1,
                 padding="same",
             )
@@ -716,7 +716,7 @@ class _TCNResidualBlock(nn.Module):
         out = self.activation(out)
         out = self.drop2(out)
 
-        out = self.reshaping_conv(out)
+        X = self.reshaping_conv(X)
 
         # ----- Residual connection -----
         out = X + out

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/braindecode/models/biot.py

@@ -17,7 +17,7 @@ class BIOT(EEGModuleMixin, nn.Module):
 
     BIOT: Cross-data Biosignal Learning in the Wild.
 
-    BIOT is a large language model for biosignal classification. It is
+    BIOT is a large brain model for biosignal classification. It is
     a wrapper around the `BIOTEncoder` and `ClassificationHead` modules.
 
     It is designed for N-dimensional biosignal data such as EEG, ECG, etc.

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/braindecode/models/eegnet.py

@@ -23,10 +23,11 @@ class EEGNet(EEGModuleMixin, nn.Sequential):
     """EEGNet model from Lawhern et al. (2018) [Lawhern2018]_.
 
     :bdg-success:`Convolution`
+
     .. figure:: https://content.cld.iop.org/journals/1741-2552/15/5/056013/revision2/jneaace8cf01_hr.jpg
-       :align: center
-       :alt: EEGNet Architecture
-       :width: 600px
+        :align: center
+        :alt: EEGNet Architecture
+        :width: 600px
 
     .. rubric:: Architectural Overview
 

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/braindecode/models/summary.csv

@@ -2,7 +2,7 @@
 ATCNet,General,Classification,250,"n_chans, n_outputs, n_times",113732,"ATCNet(n_chans=22, n_outputs=4, n_times=1000)","Convolution,Recurrent,Small Attention"
 AttentionBaseNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times",3692,"AttentionBaseNet(n_chans=22, n_outputs=4, n_times=1000)","Convolution,Small Attention"
 BDTCN,Normal Abnormal,Classification,100,"n_chans, n_outputs, n_times",456502,"BDTCN(n_chans=21, n_outputs=2, n_times=6000, n_blocks=5, n_filters=55, kernel_size=16)","Convolution,Recurrent"
-BIOT,"Sleep Staging, Epilepsy",Classification,200,"n_chans, n_outputs",3183879,"BIOT(n_chans=2,  n_outputs=5,  n_times=6000)","Large Language Model"
+BIOT,"Sleep Staging, Epilepsy",Classification,200,"n_chans, n_outputs",3183879,"BIOT(n_chans=2,  n_outputs=5,  n_times=6000)","Large Brain Model"
 ContraWR,Sleep Staging,"Classification, Embedding",125,"n_chans, n_outputs, sfreq",1160165,"ContraWR(n_chans=2, n_outputs=5, n_times=3750, emb_size=256, sfreq=125)",Convolution
 CTNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times",26900,"CTNet(n_chans=22, n_outputs=4, n_times=1000, n_filters_time=8, kernel_size=16, heads=2, emb_size=16)","Convolution,Small Attention"
 Deep4Net,General,Classification,250,"n_chans, n_outputs, n_times",282879,"Deep4Net(n_chans=22, n_outputs=4, n_times=1000)","Convolution"
@@ -16,13 +16,13 @@ EEGNeX,Motor Imagery,Classification,125,"n_chans, n_outputs, n_times",55940,"EEG
 EEGMiner,Emotion Recognition,Classification,128,"n_chans, n_outputs, n_times, sfreq",7572,"EEGMiner(n_chans=62, n_outputs=2, n_times=2560, sfreq=128)","Convolution,Interpretability"
 EEGSimpleConv,Motor Imagery,Classification,80,"n_chans, n_outputs, sfreq",730404,"EEGSimpleConv(n_chans=22,  n_outputs=4,  n_times=320, sfreq=80)","Convolution"
 EEGTCNet,Motor Imagery,Classification,250,"n_chans, n_outputs",4516,"EEGTCNet(n_chans=22,  n_outputs=4,  n_times=1000, kern_length=32)","Convolution,Recurrent"
-Labram,General,"Classification, Embedding",200,"n_chans, n_outputs, n_times",5866180,"Labram(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Convolution,Large Language Model"
+Labram,General,"Classification, Embedding",200,"n_chans, n_outputs, n_times",5866180,"Labram(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Convolution,Large Brain Model"
 MSVTNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times",75494," MSVTNet(n_chans=22,  n_outputs=4,  n_times=1000)","Convolution,Recurrent,Small Attention"
 SCCNet,Motor Imagery,Classification,125,"n_chans, n_outputs, n_times, sfreq",12070,"SCCNet(n_chans=22,  n_outputs=4,  n_times=1000, sfreq=125)","Convolution"
-SignalJEPA,"Motor Imagery, ERP, SSVEP",Embedding,128,"n_times, chs_info",3456882,"SignalJEPA(n_times=512, chs_info=Lee2019_MI().get_data(subjects=[1])[1]['0']['1train'].info[""chs""][:62])","Convolution,Channel,Large Language Model"
-SignalJEPA_Contextual,"Motor Imagery, ERP, SSVEP",Classification,128,"n_outputs, n_times, chs_info",3459184,"SignalJEPA_Contextual(n_outputs=2, input_window_seconds=4.19, sfreq=128, chs_info=Lee2019_MI().get_data(subjects=[1])[1]['0']['1train'].info[""chs""][:62])","Convolution,Channel,Large Language Model"
-SignalJEPA_PostLocal,"Motor Imagery, ERP, SSVEP",Classification,128,"n_chans, n_outputs, n_times",16142,"SignalJEPA_PostLocal(n_chans=62, n_outputs=2, input_window_seconds=4.19, sfreq=128)","Convolution,Channel,Large Language Model"
-SignalJEPA_PreLocal,"Motor Imagery, ERP, SSVEP",Classification,128,"n_outputs, n_times, chs_info",16142,"SignalJEPA_PreLocal(n_chans=62, n_outputs=2, input_window_seconds=4.19, sfreq=128)","Convolution,Channel,Large Language Model"
+SignalJEPA,"Motor Imagery, ERP, SSVEP",Embedding,128,"n_times, chs_info",3456882,"SignalJEPA(n_times=512, chs_info=Lee2019_MI().get_data(subjects=[1])[1]['0']['1train'].info[""chs""][:62])","Convolution,Channel,Large Brain Model"
+SignalJEPA_Contextual,"Motor Imagery, ERP, SSVEP",Classification,128,"n_outputs, n_times, chs_info",3459184,"SignalJEPA_Contextual(n_outputs=2, input_window_seconds=4.19, sfreq=128, chs_info=Lee2019_MI().get_data(subjects=[1])[1]['0']['1train'].info[""chs""][:62])","Convolution,Channel,Large Brain Model"
+SignalJEPA_PostLocal,"Motor Imagery, ERP, SSVEP",Classification,128,"n_chans, n_outputs, n_times",16142,"SignalJEPA_PostLocal(n_chans=62, n_outputs=2, input_window_seconds=4.19, sfreq=128)","Convolution,Channel,Large Brain Model"
+SignalJEPA_PreLocal,"Motor Imagery, ERP, SSVEP",Classification,128,"n_outputs, n_times, chs_info",16142,"SignalJEPA_PreLocal(n_chans=62, n_outputs=2, input_window_seconds=4.19, sfreq=128)","Convolution,Channel,Large Brain Model"
 SincShallowNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times, sfreq",21892,"SincShallowNet(n_chans=22,  n_outputs=4,  n_times=1000, sfreq=250)","Convolution,Interpretability"
 ShallowFBCSPNet,General,Classification,250,"n_chans, n_outputs, n_times",46084,"ShallowFBCSPNet(n_chans=22,  n_outputs=4,  n_times=1000, sfreq=250)","Convolution"
 SleepStagerBlanco2020,Sleep Staging,Classification,100,"n_chans, n_outputs, n_times",2845,"SleepStagerBlanco2020(n_chans=2, n_outputs=5, n_times=3000, sfreq=100)","Convolution"

braindecode-1.3.0.dev179592623/braindecode/version.py

@@ -0,0 +1 @@
+__version__ = "1.3.0.dev179592623"

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623/braindecode.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: braindecode
-Version: 1.3.0.dev176728557
+Version: 1.3.0.dev179592623
 Summary: Deep learning software to decode EEG, ECG or MEG signals
 Author-email: Robin Tibor Schirrmeister <robintibor@gmail.com>
 Maintainer-email: Alexandre Gramfort <agramfort@meta.com>, Bruno Aristimunha Pinto <b.aristimunha@gmail.com>, Robin Tibor Schirrmeister <robintibor@gmail.com>

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/braindecode.egg-info/SOURCES.txt

@@ -127,6 +127,6 @@ docs/models/categorization/convolution.rst
 docs/models/categorization/filterbank.rst
 docs/models/categorization/gnn.rst
 docs/models/categorization/interpretable.rst
-docs/models/categorization/llm.rst
+docs/models/categorization/lbm.rst
 docs/models/categorization/recurrent.rst
 docs/models/categorization/spd.rst

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/docs/install/install.rst

@@ -4,7 +4,7 @@
 Installation
 ================
 
-Braindecode is written in Python 3, specifically for version 3.8 or above.
+Braindecode is written in Python 3, specifically for version 3.10 or above.
 
 The package is distributed via Python package index (`PyPI <braindecode-pypi_>`_), and you can access the
 source code via `Github <braindecode-github_>`_ repository.

braindecode-1.3.0.dev176728557/docs/models/categorization/llm.rst → braindecode-1.3.0.dev179592623/docs/models/categorization/lbm.rst

@@ -4,21 +4,21 @@
 
 .. _models:
 
-|llm-icon| Large Language Models
+|lbm-icon| Large Brain Models
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 
-.. |llm-icon| image:: ../../_static/model_cat/llm.png
+.. |lbm-icon| image:: ../../_static/model_cat/lbm.png
    :height: 56px
-   :alt: Large Language Models icon
+   :alt: Large Brain Models icon
    :class: heading-icon no-scaled-link
 
-:bdg-danger:`Large Transformer Models`
+:bdg-danger:`Large Brain Models`
 
-.. figure:: ../../_static/model_connection/litmap_llm.png
+.. figure:: ../../_static/model_connection/litmap_lbm.png
     :width: 100%
     :align: center
 
-    Figure: `LitMap <https://app.litmaps.com/shared/e33fb41d-bc98-407f-9372-437f2b59a140>`__ **with large transformer model layers, last updated 26/08/2025.** Each node is a paper; rightward means more recently published, upward more cited, and links show amount of citation with logaritm scale.
+    Figure: `LitMap <https://app.litmaps.com/shared/e33fb41d-bc98-407f-9372-437f2b59a140>`__ **with large brain model layers, last updated 26/08/2025.** Each node is a paper; rightward means more recently published, upward more cited, and links show amount of citation with logaritm scale.
 
 
 .. include:: ../links.inc

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/docs/models/models_categorization.rst

@@ -21,7 +21,7 @@ How these *core* :math:`\ell` sequence transformations are structured and combin
 
 Here, we categorize the main families of brain decoding models based on their core components and design philosophies.
 The categories are not mutually exclusive, but an indication of what governs that neural network model; many models blend elements from multiple families to leverage their combined strengths.
-Beginning directly, the categories are nine: :bdg-success:`Convolution`, :bdg-secondary:`Recurrent`, :bdg-info:`Small Attention`, :bdg-primary:`Filterbank`, :bdg-warning:`Interpretability`, :bdg-danger:`Large Language Model`, :bdg-light:`Graph Neural Network`, :bdg-dark:`Symmetric Positive-Definite` and :bdg-dark-line:`Channel`.
+Beginning directly, the categories are nine: :bdg-success:`Convolution`, :bdg-secondary:`Recurrent`, :bdg-info:`Small Attention`, :bdg-primary:`Filterbank`, :bdg-warning:`Interpretability`, :bdg-danger:`Large Brain Model`, :bdg-light:`Graph Neural Network`, :bdg-dark:`Symmetric Positive-Definite` and :bdg-dark-line:`Channel`.
 
 At the moment, not all the categories are implemented, validated, and tested, but there are some that are noteworthy for introducing or popularizing concepts or layer designs that can take decoding further.
 
@@ -111,18 +111,18 @@ More details about each categories can be found in the respective sections below
 
   Learns on covariance/connectivity as SPD matrices using BiMap/ReEig/LogEig layers.
 
- .. grid-item-card:: |lightbulb| Large Transformer Models
+ .. grid-item-card:: |lightbulb| Large Brain Models
   :shadow: sm
 
-  :bdg-danger:`Large Language Model`
+  :bdg-danger:`Large Brain Model`
 
-  .. figure:: ../_static/model_cat/llm.png
+  .. figure:: ../_static/model_cat/lbm.png
    :width: 90%
    :align: center
    :alt: Diagram of transformer models
    :class: no-scaled-link
 
-  Large-scale transformer layers require self-supervised pre-training to work effectively.
+  Large-scale brain model layers require self-supervised pre-training to work effectively.
 
  .. grid-item-card:: |share-nodes| Graph Neural Network
   :shadow: sm
@@ -160,7 +160,7 @@ More details about each categories can be found in the respective sections below
 - **Filterbank-style models** (:bdg-primary:`Filterbank`) explicitly decompose signals into multiple bands before (or while) learning, echoing the classic FBCSP pipeline; examples include :class:`FBCNet` and :class:`FBMSNet` (:cite:label:`mane2021fbcnet,liu2022fbmsnet`).
 - **Interpretability-by-design** (:bdg-warning:`Interpretability`) architectures expose physiologically meaningful primitives (e.g., band-pass/sinc filters, variance or connectivity features), enabling direct neuroscientific inspection; see :class:`SincShallowNet` and :class:`EEGMiner` (:cite:label:`borra2020interpretable,ludwig2024eegminer`).
 - **SPD / Riemannian** (:bdg-dark:`SPD`) methods operate on covariance (or connectivity) matrices as points on the SPD manifold, combining layers such as BiMap, ReEig, and LogEig; deep SPD networks and Riemannian classifiers motivate this family (:cite:label:`huang2017riemannian`). *(Coming soon in a dedicate repository.)*
-- **Large-model / Transformer** (:bdg-danger:`Large Language Model`) approaches pretrain attention-based encoders on diverse biosignals and fine-tune for EEG tasks; e.g., :class:`BIOT` (:cite:label:`yang2023biot`). These typically need a heavily self-supervised pre-training before decoding.
+- **Large-model / Transformer** (:bdg-danger:`Large Brain Model`) approaches pretrain attention-based encoders on diverse biosignals and fine-tune for EEG tasks; e.g., :class:`BIOT` (:cite:label:`yang2023biot`), :class:`Labram` (:cite:label:`jiang2024large`). These typically need a heavily self-supervised pre-training before decoding.
 - **Graph neural networks** (:bdg-light:`Graph Neural Network`) treat channels/regions as nodes with learned (static or dynamic) edges to model functional connectivity explicitly; representative EEG-GNN, more common in the epileptic decoding (:cite:label:`klepl2024graph`).
 - **Channel-domain robustness** (:bdg-dark-line:`Channel`) techniques target variability in electrode layouts by learning montage-agnostic or channel-selective layers (e.g., dynamic spatial filtering, differentiable channel re-ordering); these strategies improve cross-setup generalization :class:`SignalJEPA` (:cite:label:`guetschel2024sjepa,chen2024eegprogress`).
 
@@ -194,7 +194,7 @@ We appreciate your contributions to expanding the library!
    categorization/filterbank
    categorization/interpretable
    categorization/spd
-   categorization/llm
+   categorization/lbm
    categorization/gnn
    categorization/channel
 

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/docs/models/models_table.rst

@@ -19,7 +19,7 @@
   .tag.tag-filterbank{ background:#06b6d4!important; color:#073042!important; border-color:#0891b2!important; }
   .tag.tag-interp    { background:#f59e0b!important; color:#1f2937!important; border-color:#d97706!important; }
   .tag.tag-spd       { background:#111827!important; color:#fff!important;  border-color:#0b1220!important; }
-  .tag.tag-llm       { background:#e11d48!important; color:#fff!important;  border-color:#be123c!important; }
+  .tag.tag-lbm       { background:#e11d48!important; color:#fff!important;  border-color:#be123c!important; }
   .tag.tag-gnn       { background:#475569!important; color:#fff!important;  border-color:#334155!important; }
   .tag.tag-channel   { background:#64748b!important; color:#fff!important;  border-color:#475569!important; }
 
@@ -83,7 +83,7 @@ We are continually expanding this collection and welcome contributions! If you h
   .tag.tag-filterbank{ background:#06b6d4!important; color:#073042!important; border-color:#0891b2!important; }
   .tag.tag-interp    { background:#f59e0b!important; color:#1f2937!important; border-color:#d97706!important; }
   .tag.tag-spd       { background:#111827!important; color:#fff!important;  border-color:#0b1220!important; }
-  .tag.tag-llm       { background:#e11d48!important; color:#fff!important;  border-color:#be123c!important; }
+  .tag.tag-lbm       { background:#e11d48!important; color:#fff!important;  border-color:#be123c!important; }
   .tag.tag-gnn       { background:#475569!important; color:#fff!important;  border-color:#334155!important; }
   .tag.tag-channel   { background:#64748b!important; color:#fff!important;  border-color:#475569!important; }
   .tag { box-shadow:0 0 0 0 rgba(0,0,0,0); transition:box-shadow .15s, transform .15s; }
@@ -111,11 +111,11 @@ We are continually expanding this collection and welcome contributions! If you h
         "Convolution":"tag-conv","Recurrent":"tag-recurrent","Small Attention":"tag-smallattn",
         "Filterbank":"tag-filterbank","Interpretability":"tag-interp",
         "SPD":"tag-spd","Riemannian":"tag-spd",
-        "Large Language Model":"tag-llm","Graph Neural Network":"tag-gnn","Channel":"tag-channel"
+        "Large Brain Model":"tag-lbm","Graph Neural Network":"tag-gnn","Channel":"tag-channel"
       };
       $(ctx).find('.tag').each(function () {
         var t = $(this).text().trim();
-        $(this).removeClass('tag-conv tag-recurrent tag-smallattn tag-filterbank tag-interp tag-spd tag-llm tag-gnn tag-channel');
+        $(this).removeClass('tag-conv tag-recurrent tag-smallattn tag-filterbank tag-interp tag-spd tag-lbm tag-gnn tag-channel');
         if (map[t]) $(this).addClass(map[t]);
         else if (t.includes('SPD') || t.includes('Riemannian')) $(this).addClass('tag-spd');
       });

{braindecode-1.3.0.dev176728557 → braindecode-1.3.0.dev179592623}/docs/whats_new.rst

@@ -28,6 +28,7 @@ API changes
 
 Bugs
 ~~~~
+- Fixing the issues with :class:`braindecode.models.ATCNet` (:gh:`782` by `Hamdi Altaheri`_)
 
 
 Current 1.2 (stable)
@@ -396,3 +397,4 @@ Authors
 .. _Young Truong: https://github.com/dungscout96
 .. _Lucas Heck: https://github.com/lucas-heck
 .. _Ganasekhar Kalla: https://github.com/Ganasekhar-gif
+.. _Hamdi Altaheri: https://github.com/Altaheri

braindecode-1.3.0.dev176728557/braindecode/version.py

@@ -1 +0,0 @@
-__version__ = "1.3.0.dev176728557"