braindecode 1.3.0.dev181065563__py3-none-any.whl → 1.3.0.dev183667303__py3-none-any.whl

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"
@@ -37,3 +37,6 @@ FBCNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times, sfreq",118
 FBMSNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times, sfreq",16231,"FBMSNet(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Convolution,FilterBank"
 FBLightConvNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times, sfreq",6596,"FBLightConvNet(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Convolution,FilterBank"
 IFNet,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times, sfreq",9860,"IFNet(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Convolution,FilterBank"
+PBT,General,Classification,250,"n_chans, n_outputs, n_times",818948,"PBT(n_chans=22, n_outputs=4, n_times=1000, sfreq=250)","Large Brain Model"
+SSTDPN,Motor Imagery,Classification,250,"n_chans, n_outputs, n_times",19502,"SSTDPN(n_chans=22, n_outputs=4, n_times=1000)","Convolution,Small Attention"
+BENDR,General,"Classification,Embedding",250,"n_chans, n_times, n_outputs",157141049,"BENDR(n_chans=22, n_outputs=4, n_times=1000)","Large Brain Model,Convolution"

braindecode/models/usleep.py

@@ -62,43 +62,48 @@ class USleep(EEGModuleMixin, nn.Module):
     - Decoder :class:`_DecoderBlock`  **(progressive upsampling + skip fusion to high-frequency map, 12 blocks; upsampling x2 per block)**
 
         - *Operations.*
-        - **Nearest-neighbor upsample**, :class:`nn.Upsample` (x2)
-        - **Convolution2d** (k=2), :class:`torch.nn.Conv2d`
-        - ELU, :class:`torch.nn.ELU`
-        - Batch Norm, :class:`torch.nn.BatchNorm2d`
-        - **Concatenate** with the encoder skip at the same temporal scale, :function:`torch.cat`
-        - **Convolution**, :class:`torch.nn.Conv2d`
-        - ELU, :class:`torch.nn.ELU`
-        - Batch Norm, :class:`torch.nn.BatchNorm2d`.
+
+            - **Nearest-neighbor upsample**, :class:`nn.Upsample` (x2)
+            - **Convolution2d** (k=2), :class:`torch.nn.Conv2d`
+            - ELU, :class:`torch.nn.ELU`
+            - Batch Norm, :class:`torch.nn.BatchNorm2d`
+            - **Concatenate** with the encoder skip at the same temporal scale, ``torch.cat``
+            - **Convolution**, :class:`torch.nn.Conv2d`
+            - ELU, :class:`torch.nn.ELU`
+            - Batch Norm, :class:`torch.nn.BatchNorm2d`.
 
     **Output**: A multi-class, **high-frequency** per-sample representation aligned to the input rate (128 Hz).
 
     - **Segment Classifier incorporate into :class:`braindecode.models.USleep` (aggregation to fixed epochs)**
 
         - *Operations.*
-        - **Mean-pool**, :class:`torch.nn.AvgPool2d` per class with kernel = epoch length *i* and stride *i*
-        - **1x1 conv**, :class:`torch.nn.Conv2d`
-        - ELU, :class:`torch.nn.ELU`
-        - **1x1 conv**, :class:`torch.nn.Conv2d` with ``(T, K)`` (epochs x stages).
+
+            - **Mean-pool**, :class:`torch.nn.AvgPool2d` per class with kernel = epoch length *i* and stride *i*
+            - **1x1 conv**, :class:`torch.nn.Conv2d`
+            - ELU, :class:`torch.nn.ELU`
+            - **1x1 conv**, :class:`torch.nn.Conv2d` with ``(T, K)`` (epochs x stages).
 
     **Role**: Learns a **non-linear** weighted combination over each 30-s window (unlike U-Time's linear combiner).
 
     .. rubric:: Convolutional Details
 
     - **Temporal (where time-domain patterns are learned).**
-    All convolutions are **1-D along time**; depth (12 levels) plus pooling yields an extensive receptive field
-    (reported sensitivity to ±6.75 min around each epoch; theoretical field ≈ 9.6 min at the deepest layer).
-    The decoder restores sample-level resolution before epoch aggregation.
+
+      All convolutions are **1-D along time**; depth (12 levels) plus pooling yields an extensive receptive field
+      (reported sensitivity to ±6.75 min around each epoch; theoretical field ≈ 9.6 min at the deepest layer).
+      The decoder restores sample-level resolution before epoch aggregation.
 
     - **Spatial (how channels are processed).**
-    Convolutions mix across the *channel* dimension jointly with time (no separate spatial operator). The system
-    is **montage-agnostic** (any reasonable EEG/EOG pair) and was trained across diverse cohorts/protocols,
-    supporting robustness to channel placement and hardware differences.
+
+      Convolutions mix across the *channel* dimension jointly with time (no separate spatial operator). The system
+      is **montage-agnostic** (any reasonable EEG/EOG pair) and was trained across diverse cohorts/protocols,
+      supporting robustness to channel placement and hardware differences.
 
     - **Spectral (how frequency content is captured).**
-    No explicit Fourier/wavelet transform is used; the **stack of temporal convolutions** acts as a learned
-    filter bank whose effective bandwidth grows with depth. The high-frequency decoder output (128 Hz)
-    retains fine temporal detail for the segment classifier.
+
+      No explicit Fourier/wavelet transform is used; the **stack of temporal convolutions** acts as a learned
+      filter bank whose effective bandwidth grows with depth. The high-frequency decoder output (128 Hz)
+      retains fine temporal detail for the segment classifier.
 
 
     .. rubric:: Attention / Sequential Modules

braindecode/models/util.py

@@ -95,6 +95,9 @@ models_mandatory_parameters = [
     ("FBMSNet", ["n_chans", "n_outputs", "n_times", "sfreq"], dict(sfreq=200.0)),
     ("FBLightConvNet", ["n_chans", "n_outputs", "n_times", "sfreq"], dict(sfreq=200.0)),
     ("IFNet", ["n_chans", "n_outputs", "n_times", "sfreq"], dict(sfreq=200.0)),
+    ("PBT", ["n_chans", "n_outputs", "n_times"], None),
+    ("SSTDPN", ["n_chans", "n_outputs", "n_times", "sfreq"], None),
+    ("BENDR", ["n_chans", "n_outputs", "n_times"], None),
 ]
 
 ################################################################

braindecode/modules/attention.py

@@ -38,7 +38,7 @@ class SqueezeAndExcitation(nn.Module):
     References
     ----------
     .. [Hu2018] Hu, J., Albanie, S., Sun, G., Wu, E., 2018.
-    Squeeze-and-Excitation Networks. CVPR 2018.
+        Squeeze-and-Excitation Networks. CVPR 2018.
     """
 
     def __init__(self, in_channels: int, reduction_rate: int, bias: bool = False):
@@ -93,7 +93,7 @@ class GSoP(nn.Module):
     References
     ----------
     .. [Gao2018] Gao, Z., Jiangtao, X., Wang, Q., Li, P., 2018.
-    Global Second-order Pooling Convolutional Networks. CVPR 2018.
+        Global Second-order Pooling Convolutional Networks. CVPR 2018.
     """
 
     def __init__(self, in_channels: int, reduction_rate: int, bias: bool = True):
@@ -149,7 +149,7 @@ class FCA(nn.Module):
     References
     ----------
     .. [Qin2021] Qin, Z., Zhang, P., Wu, F., Li, X., 2021.
-    FcaNet: Frequency Channel Attention Networks. ICCV 2021.
+        FcaNet: Frequency Channel Attention Networks. ICCV 2021.
     """
 
     def __init__(
@@ -233,7 +233,7 @@ class EncNet(nn.Module):
     References
     ----------
     .. [Zhang2018] Zhang, H. et al. 2018.
-    Context Encoding for Semantic Segmentation. CVPR 2018.
+        Context Encoding for Semantic Segmentation. CVPR 2018.
     """
 
     def __init__(self, in_channels: int, n_codewords: int):
@@ -290,7 +290,7 @@ class ECA(nn.Module):
     References
     ----------
     .. [Wang2021] Wang, Q. et al., 2021. ECA-Net: Efficient Channel Attention
-    for Deep Convolutional Neural Networks. CVPR 2021.
+        for Deep Convolutional Neural Networks. CVPR 2021.
     """
 
     def __init__(self, in_channels: int, kernel_size: int):
@@ -341,8 +341,8 @@ class GatherExcite(nn.Module):
     References
     ----------
     .. [Hu2018b] Hu, J., Albanie, S., Sun, G., Vedaldi, A., 2018.
-    Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks.
-    NeurIPS 2018.
+        Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks.
+        NeurIPS 2018.
     """
 
     def __init__(
@@ -410,7 +410,7 @@ class GCT(nn.Module):
     References
     ----------
     .. [Yang2020] Yang, Z. Linchao, Z., Wu, Y., Yang, Y., 2020.
-    Gated Channel Transformation for Visual Recognition. CVPR 2020.
+        Gated Channel Transformation for Visual Recognition. CVPR 2020.
     """
 
     def __init__(self, in_channels: int):
@@ -455,7 +455,7 @@ class SRM(nn.Module):
     References
     ----------
     .. [Lee2019] Lee, H., Kim, H., Nam, H., 2019. SRM: A Style-based
-    Recalibration Module for Convolutional Neural Networks. ICCV 2019.
+        Recalibration Module for Convolutional Neural Networks. ICCV 2019.
     """
 
     def __init__(
@@ -520,7 +520,7 @@ class CBAM(nn.Module):
     References
     ----------
     .. [Woo2018] Woo, S., Park, J., Lee, J., Kweon, I., 2018.
-    CBAM: Convolutional Block Attention Module. ECCV 2018.
+        CBAM: Convolutional Block Attention Module. ECCV 2018.
     """
 
     def __init__(self, in_channels: int, reduction_rate: int, kernel_size: int):

braindecode/modules/blocks.py

@@ -37,8 +37,8 @@ class MLP(nn.Sequential):
     :math:`a_i` are called activation functions. The trainable parameters of an
     MLP are its weights and biases :math:`\\phi = \{W_i, b_i | i = 1, \dots, L\}`.
 
-    Parameters:
-    -----------
+    Parameters
+    ----------
     in_features: int
         Number of input features.
     hidden_features: Sequential[int] (default=None)
@@ -49,7 +49,7 @@ class MLP(nn.Sequential):
     out_features: int (default=None)
         Number of output features, if None, set to in_features.
     act_layer: nn.GELU (default)
-        The activation function constructor. If :py:`None`, use
+        The activation function constructor. If ``None``, use
         :class:`torch.nn.GELU` instead.
     drop: float (default=0.0)
         Dropout rate.

braindecode/modules/filter.py

@@ -17,9 +17,8 @@ class FilterBankLayer(nn.Module):
     It uses MNE's `create_filter` function to create the band-specific filters and
     applies them to multi-channel time-series data. Each filter in the bank corresponds to a
     specific frequency band and is applied to all channels of the input data. The filtering is
-    performed using FFT-based convolution via the `fftconvolve` function from
-    :func:`torchaudio.functional if the method is FIR, and `filtfilt` function from
-    :func:`torchaudio.functional if the method is IIR.
+    performed using FFT-based convolution via the ``torchaudio.functional`` if the method is FIR,
+    and ``torchaudio.functional`` if the method is IIR.
 
     The default configuration creates 9 non-overlapping frequency bands with a 4 Hz bandwidth,
     spanning from 4 Hz to 40 Hz (i.e., 4-8 Hz, 8-12 Hz, ..., 36-40 Hz). This setup is based on the

braindecode/modules/layers.py

@@ -70,26 +70,27 @@ class TimeDistributed(nn.Module):
 class DropPath(nn.Module):
     """Drop paths, also known as Stochastic Depth, per sample.
 
-        When applied in main path of residual blocks.
+    When applied in main path of residual blocks.
 
-        Parameters:
-        -----------
-        drop_prob: float (default=None)
-            Drop path probability (should be in range 0-1).
+    Parameters
+    ----------
+    drop_prob: float (default=None)
+        Drop path probability (should be in range 0-1).
 
-        Notes
-        -----
-        Code copied and modified from VISSL facebookresearch:
+    Notes
+    -----
+    Code copied and modified from VISSL facebookresearch:
     https://github.com/facebookresearch/vissl/blob/0b5d6a94437bc00baed112ca90c9d78c6ccfbafb/vissl/models/model_helpers.py#L676
-        All rights reserved.
-
-        THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
-        IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
-        FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
-        AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
-        LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
-        OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
-        SOFTWARE.
+
+    All rights reserved.
+
+    THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+    IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+    FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+    AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+    LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+    OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+    SOFTWARE.
     """
 
     def __init__(self, drop_prob=None):

braindecode/preprocessing/__init__.py

@@ -1,3 +1,16 @@
+from .eegprep_preprocess import (
+    EEGPrep,
+    ReinterpolateRemovedChannels,
+    RemoveBadChannels,
+    RemoveBadChannelsNoLocs,
+    RemoveBadWindows,
+    RemoveBursts,
+    RemoveCommonAverageReference,
+    RemoveDCOffset,
+    RemoveDrifts,
+    RemoveFlatChannels,
+    Resampling,
+)
 from .mne_preprocess import (  # type: ignore[attr-defined]
     Crop,
     DropChannels,
@@ -31,6 +44,17 @@ __all__ = [
     "Filter",
     "Pick",
     "Crop",
+    "EEGPrep",
+    "RemoveDCOffset",
+    "Resampling",
+    "RemoveFlatChannels",
+    "RemoveDrifts",
+    "RemoveBadChannels",
+    "RemoveBadChannelsNoLocs",
+    "RemoveBursts",
+    "RemoveBadWindows",
+    "ReinterpolateRemovedChannels",
+    "RemoveCommonAverageReference",
     "create_windows_from_events",
     "create_fixed_length_windows",
     "create_windows_from_target_channels",