braindecode 0.8.1__py3-none-any.whl → 1.1.0__py3-none-any.whl

braindecode/models/modules.py

@@ -1,358 +0,0 @@
-# Authors: Robin Schirrmeister <robintibor@gmail.com>
-#
-# License: BSD (3-clause)
-
-import numpy as np
-
-import torch
-from torch import nn
-import torch.nn.functional as F
-
-from ..util import np_to_th
-
-
-class Ensure4d(nn.Module):
-    def forward(self, x):
-        while len(x.shape) < 4:
-            x = x.unsqueeze(-1)
-        return x
-
-
-class Expression(nn.Module):
-    """Compute given expression on forward pass.
-
-    Parameters
-    ----------
-    expression_fn : callable
-        Should accept variable number of objects of type
-        `torch.autograd.Variable` to compute its output.
-    """
-
-    def __init__(self, expression_fn):
-        super(Expression, self).__init__()
-        self.expression_fn = expression_fn
-
-    def forward(self, *x):
-        return self.expression_fn(*x)
-
-    def __repr__(self):
-        if hasattr(self.expression_fn, "func") and hasattr(
-            self.expression_fn, "kwargs"
-        ):
-            expression_str = "{:s} {:s}".format(
-                self.expression_fn.func.__name__, str(self.expression_fn.kwargs)
-            )
-        elif hasattr(self.expression_fn, "__name__"):
-            expression_str = self.expression_fn.__name__
-        else:
-            expression_str = repr(self.expression_fn)
-        return self.__class__.__name__ + "(expression=%s) " % expression_str
-
-
-class AvgPool2dWithConv(nn.Module):
-    """
-    Compute average pooling using a convolution, to have the dilation parameter.
-
-    Parameters
-    ----------
-    kernel_size: (int,int)
-        Size of the pooling region.
-    stride: (int,int)
-        Stride of the pooling operation.
-    dilation: int or (int,int)
-        Dilation applied to the pooling filter.
-    padding: int or (int,int)
-        Padding applied before the pooling operation.
-    """
-
-    def __init__(self, kernel_size, stride, dilation=1, padding=0):
-        super(AvgPool2dWithConv, self).__init__()
-        self.kernel_size = kernel_size
-        self.stride = stride
-        self.dilation = dilation
-        self.padding = padding
-        # don't name them "weights" to
-        # make sure these are not accidentally used by some procedure
-        # that initializes parameters or something
-        self._pool_weights = None
-
-    def forward(self, x):
-        # Create weights for the convolution on demand:
-        # size or type of x changed...
-        in_channels = x.size()[1]
-        weight_shape = (
-            in_channels,
-            1,
-            self.kernel_size[0],
-            self.kernel_size[1],
-        )
-        if self._pool_weights is None or (
-            (tuple(self._pool_weights.size()) != tuple(weight_shape))
-            or (self._pool_weights.is_cuda != x.is_cuda)
-            or (self._pool_weights.data.type() != x.data.type())
-        ):
-            n_pool = np.prod(self.kernel_size)
-            weights = np_to_th(np.ones(weight_shape, dtype=np.float32) / float(n_pool))
-            weights = weights.type_as(x)
-            if x.is_cuda:
-                weights = weights.cuda()
-            self._pool_weights = weights
-
-        pooled = F.conv2d(
-            x,
-            self._pool_weights,
-            bias=None,
-            stride=self.stride,
-            dilation=self.dilation,
-            padding=self.padding,
-            groups=in_channels,
-        )
-        return pooled
-
-
-class IntermediateOutputWrapper(nn.Module):
-    """Wraps network model such that outputs of intermediate layers can be returned.
-    forward() returns list of intermediate activations in a network during forward pass.
-
-    Parameters
-    ----------
-    to_select : list
-        list of module names for which activation should be returned
-    model : model object
-        network model
-
-    Examples
-    --------
-    >>> model = Deep4Net()
-    >>> select_modules = ['conv_spat','conv_2','conv_3','conv_4'] # Specify intermediate outputs
-    >>> model_pert = IntermediateOutputWrapper(select_modules,model) # Wrap model
-    """
-
-    def __init__(self, to_select, model):
-        if not len(list(model.children())) == len(list(model.named_children())):
-            raise Exception("All modules in model need to have names!")
-
-        super(IntermediateOutputWrapper, self).__init__()
-
-        modules_list = model.named_children()
-        for key, module in modules_list:
-            self.add_module(key, module)
-            self._modules[key].load_state_dict(module.state_dict())
-        self._to_select = to_select
-
-    def forward(self, x):
-        # Call modules individually and append activation to output if module is in to_select
-        o = []
-        for name, module in self._modules.items():
-            x = module(x)
-            if name in self._to_select:
-                o.append(x)
-        return o
-
-
-class TimeDistributed(nn.Module):
-    """Apply module on multiple windows.
-
-    Apply the provided module on a sequence of windows and return their
-    concatenation.
-    Useful with sequence-to-prediction models (e.g. sleep stager which must map
-    a sequence of consecutive windows to the label of the middle window in the
-    sequence).
-
-    Parameters
-    ----------
-    module : nn.Module
-        Module to be applied to the input windows. Must accept an input of
-        shape (batch_size, n_channels, n_times).
-    """
-
-    def __init__(self, module):
-        super().__init__()
-        self.module = module
-
-    def forward(self, x):
-        """
-        Parameters
-        ----------
-        x : torch.Tensor
-            Sequence of windows, of shape (batch_size, seq_len, n_channels,
-            n_times).
-
-        Returns
-        -------
-        torch.Tensor
-            Shape (batch_size, seq_len, output_size).
-        """
-        b, s, c, t = x.shape
-        out = self.module(x.view(b * s, c, t))
-        return out.view(b, s, -1)
-
-
-class CausalConv1d(nn.Conv1d):
-    """Causal 1-dimensional convolution
-
-    Code modified from [1]_ and [2]_.
-
-    Parameters
-    ----------
-    in_channels : int
-        Input channels.
-    out_channels : int
-        Output channels (number of filters).
-    kernel_size : int
-        Kernel size.
-    dilation : int, optional
-        Dilation (number of elements to skip within kernel multiplication).
-        Default to 1.
-    **kwargs :
-        Other keyword arguments to pass to torch.nn.Conv1d, except for
-        `padding`!!
-
-    References
-    ----------
-    .. [1] https://discuss.pytorch.org/t/causal-convolution/3456/4
-    .. [2] https://gist.github.com/paultsw/7a9d6e3ce7b70e9e2c61bc9287addefc
-    """
-
-    def __init__(
-        self,
-        in_channels,
-        out_channels,
-        kernel_size,
-        dilation=1,
-        **kwargs,
-    ):
-        assert "padding" not in kwargs, (
-            "The padding parameter is controlled internally by "
-            f"{type(self).__name__} class. You should not try to override this"
-            " parameter."
-        )
-
-        super().__init__(
-            in_channels=in_channels,
-            out_channels=out_channels,
-            kernel_size=kernel_size,
-            dilation=dilation,
-            padding=(kernel_size - 1) * dilation,
-            **kwargs,
-        )
-
-    def forward(self, X):
-        out = super().forward(X)
-        return out[..., : -self.padding[0]]
-
-
-class MaxNormLinear(nn.Linear):
-    """Linear layer with MaxNorm constraining on weights.
-
-    Equivalent of Keras tf.keras.Dense(..., kernel_constraint=max_norm())
-    [1, 2]_. Implemented as advised in [3]_.
-
-    Parameters
-    ----------
-    in_features: int
-        Size of each input sample.
-    out_features: int
-        Size of each output sample.
-    bias: bool, optional
-        If set to ``False``, the layer will not learn an additive bias.
-        Default: ``True``.
-
-    References
-    ----------
-    .. [1] https://keras.io/api/layers/core_layers/dense/#dense-class
-    .. [2] https://www.tensorflow.org/api_docs/python/tf/keras/constraints/
-           MaxNorm
-    .. [3] https://discuss.pytorch.org/t/how-to-correctly-implement-in-place-
-           max-norm-constraint/96769
-    """
-
-    def __init__(
-        self, in_features, out_features, bias=True, max_norm_val=2, eps=1e-5, **kwargs
-    ):
-        super().__init__(
-            in_features=in_features, out_features=out_features, bias=bias, **kwargs
-        )
-        self._max_norm_val = max_norm_val
-        self._eps = eps
-
-    def forward(self, X):
-        self._max_norm()
-        return super().forward(X)
-
-    def _max_norm(self):
-        with torch.no_grad():
-            norm = self.weight.norm(2, dim=0, keepdim=True).clamp(
-                min=self._max_norm_val / 2
-            )
-            desired = torch.clamp(norm, max=self._max_norm_val)
-            self.weight *= desired / (self._eps + norm)
-
-
-class CombinedConv(nn.Module):
-    """Merged convolutional layer for temporal and spatial convs in Deep4/ShallowFBCSP
-
-    Numerically equivalent to the separate sequential approach, but this should be faster.
-
-    Parameters
-    ----------
-    in_chans : int
-        Number of EEG input channels.
-    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.
-    bias_time: bool
-        Whether to use bias in the temporal conv
-    bias_spat: bool
-        Whether to use bias in the spatial conv
-
-    """
-
-    def __init__(
-        self,
-        in_chans,
-        n_filters_time=40,
-        n_filters_spat=40,
-        filter_time_length=25,
-        bias_time=True,
-        bias_spat=True,
-    ):
-        super().__init__()
-        self.bias_time = bias_time
-        self.bias_spat = bias_spat
-        self.conv_time = nn.Conv2d(
-            1, n_filters_time, (filter_time_length, 1), bias=bias_time, stride=1
-        )
-        self.conv_spat = nn.Conv2d(
-            n_filters_time, n_filters_spat, (1, in_chans), bias=bias_spat, stride=1
-        )
-
-    def forward(self, x):
-        # Merge time and spat weights
-        combined_weight = (
-            (self.conv_time.weight * self.conv_spat.weight.permute(1, 0, 2, 3))
-            .sum(0)
-            .unsqueeze(1)
-        )
-
-        # Calculate bias term
-        if not self.bias_spat and not self.bias_time:
-            bias = None
-        else:
-            bias = 0
-            if self.bias_time:
-                bias += (
-                    self.conv_spat.weight.squeeze()
-                    .sum(-1)
-                    .mm(self.conv_time.bias.unsqueeze(-1))
-                    .squeeze()
-                )
-            if self.bias_spat:
-                bias += self.conv_spat.bias
-
-        return F.conv2d(
-            x, weight=combined_weight, bias=bias, stride=(1, 1)
-        )

braindecode-0.8.1.dist-info/RECORD

@@ -1,68 +0,0 @@
-braindecode/__init__.py,sha256=DVr7xakvqGSkno_VZoOmXHot-PeRHfkSGGBwGq8b1Us,183
-braindecode/classifier.py,sha256=IONNZmZHyoKmWSdmuL9pZqEDxlS0aJvcLD_M9O41CP0,9816
-braindecode/eegneuralnet.py,sha256=P8Trs2m6X6lEKnrQH-FOdQPmQ8cqaLwPl1rEzk_pJaw,14818
-braindecode/regressor.py,sha256=zXrYi3fEvQdROe-BQ79ZFHfbrCvFb4KJa0_zesbvAXw,9359
-braindecode/util.py,sha256=OW4UATTHvnpem9O6vD3UfJSA0y76kons_VBchVPMGM8,13037
-braindecode/version.py,sha256=WY9HbkgoxJlUCfEaMASIAo3wFTh3K3UXHty6LsjolMg,20
-braindecode/augmentation/__init__.py,sha256=WmtUAQVKR735X1B9e9Dz1_nFUUVs2jlmF0p9yOaLLsA,869
-braindecode/augmentation/base.py,sha256=MvUnxJcXdsu_uKfSHFzVVQ_0bal6QCc1dBndLsP62fw,7356
-braindecode/augmentation/functional.py,sha256=QTqlRMiVLdmwM_pJacZXvyTE6FT47pPmpcZ3TohIGpc,33343
-braindecode/augmentation/transforms.py,sha256=UZj0C7ksJmbs6PD302jAXVRbkTDsW28_CgP1SS0KFvg,38000
-braindecode/datasets/__init__.py,sha256=fsIBanNucO9xCKKtMKxI1aDgg6nYoE9amnMAI7iZXx8,658
-braindecode/datasets/base.py,sha256=HHBICxUjNS2Jz80i08ZPwyaQimM-DCqNGPEc55uqz2A,30613
-braindecode/datasets/bbci.py,sha256=eeX408xg_VtgxsIoev21VJXbyatjQMDO4WRsJtbpsks,19130
-braindecode/datasets/bcicomp.py,sha256=cJZiF2Z6IB0nlkFXbtCftke8zxoV8Ics24nL48Ed7uU,7421
-braindecode/datasets/mne.py,sha256=bcR2vNWuRgE1KjFIheX-v5uvO2nQDJFcVXgVDC1zbeA,5824
-braindecode/datasets/moabb.py,sha256=qHC_4VBJw1A5OCJt_vfHvZK9FSKv_nSoIvS2N6AZE1M,5134
-braindecode/datasets/sleep_physionet.py,sha256=m1twC8MxQweyx0zBK0ZrrBP9EmyIdqvGmFboU6viz_I,3908
-braindecode/datasets/tuh.py,sha256=xFjn7C50kZ8npHCrB0unjZ_EghvlGNfrWogNHsuxAoo,18450
-braindecode/datasets/xy.py,sha256=n2d_rYyLJZEyrLiqJlUhxMtU44Ok2feAfZRqGnQP47k,2799
-braindecode/datautil/__init__.py,sha256=G6E1sIxPr2awx8UvvJG3FNxqRR-OHuJGWYIb_vuKjFU,1628
-braindecode/datautil/mne.py,sha256=lSaXS9ZwYY11JaOVRvP0q7buoLLbr4xWZ7SCe_y_u4E,287
-braindecode/datautil/preprocess.py,sha256=jwO58bP0-NU2AczFJUrIRIOicpbCOymsabhm-UNQ_9o,374
-braindecode/datautil/serialization.py,sha256=WyVdjubnZbKZ9FHQBZECKyfD6_m2kGC6tySEk3RmcSI,12430
-braindecode/datautil/util.py,sha256=ZfDoxLieKsgI8xcWQqebV-vJ5pJYRvRRHkEwhwpgoKU,674
-braindecode/datautil/windowers.py,sha256=0QsJiOWvdWxoTFKceVXxqmWoNyVKtpBxk8VllpTysFM,251
-braindecode/datautil/xy.py,sha256=rS4W19sAKG94pjZEQLqYXlvy2h8IInkoDYM0QlixzHs,283
-braindecode/models/__init__.py,sha256=epIddo-wqmWEPgT3ONHh5fhAGr4l9y0XMpinmi1iVYA,1059
-braindecode/models/atcnet.py,sha256=0ckyrqTmeGAwNBcCZXedDpyyZS8a--g3RS7I6yjQT0U,23059
-braindecode/models/base.py,sha256=O23DV08ef-QFM-h8jkKRUt5OLZ8tQPa7kvFUy3a9uGk,11158
-braindecode/models/deep4.py,sha256=WABnxH2iN_PMNYunqLnduxB5cO8WPHufWTcu_nEBLKA,12675
-braindecode/models/deepsleepnet.py,sha256=l5xT9Ri7_cLPzdInrTpctEgtncWV62mCssW24W2aln8,7821
-braindecode/models/eegconformer.py,sha256=Iae5GARAELxcYHiZsbFfKqbVTnH1QhZ9AvlM8xuNaYg,14228
-braindecode/models/eeginception.py,sha256=P6qzqkf8ug-c4kLLeOqiJAgSKbuoAOh-onn6pwDFwLo,12148
-braindecode/models/eeginception_erp.py,sha256=xf1HdPF14bAlAIUkmdB29gn7katPw2dv6UcwGuBKUW4,12150
-braindecode/models/eeginception_mi.py,sha256=yvLEZwVAduIMnEvZN6C5PLOecd5LDCrnGO0FZkhuNp8,10219
-braindecode/models/eegitnet.py,sha256=TfH5RWU_eEUO0gDp-rgvUEl1Y7ehE0oykUm3PovgnCo,7858
-braindecode/models/eegnet.py,sha256=08DBm4cFYcVZT7oZrP97-pMeYQ3Yya97UQsEKPdO0IY,13478
-braindecode/models/eegresnet.py,sha256=pyMA711K03_Hi6TCkddfJb46j6S68ekKTCQ3uelaXYM,11655
-braindecode/models/functions.py,sha256=_GfkZHSWR2X2QNH_7-KeU8TPIO8Mu6uC-RdZEWmuOEM,917
-braindecode/models/hybrid.py,sha256=rQCZIXecsP7phxUdM8h64b4pld-eApz-pu6DN7bTgeo,4865
-braindecode/models/modules.py,sha256=p4J4I4PcJWpyUaAOSGSai10-PrpBR3S2Vyqt50zzUqA,10672
-braindecode/models/shallow_fbcsp.py,sha256=zMLTokGwSSp8K_aSqc7TaQBs_hkM0IbuBq7-F8_QXkY,8078
-braindecode/models/sleep_stager_blanco_2020.py,sha256=fqsv9W9ba_3ZkHAgnPRFCUjk-30pUOObXEumgFsbcVo,5920
-braindecode/models/sleep_stager_chambon_2018.py,sha256=h_EcnoFf1hhdl-YI5jrgRBW7Lv1nGdAOsYwsrmo_vwY,5314
-braindecode/models/sleep_stager_eldele_2021.py,sha256=jnwF3MXH1xbCcNrDrVS0A9Xoy0w76LLf_td2lZShBzY,15141
-braindecode/models/tcn.py,sha256=KrjhwgGN-bzrMGlxZ6Iz6uDKINiZQGhFkGfrHpNA-GE,6389
-braindecode/models/tidnet.py,sha256=TrnS-uYYcYS2lxTH0m-TZOHIl57nh0XAtA8wzigOVZw,9886
-braindecode/models/usleep.py,sha256=ILeqiBvse9Weg_neL_uO34H5-MMlD53ijdtGmFX2kDA,11465
-braindecode/models/util.py,sha256=ntEkLSyFDjBC2X47YLOdNfJxLYc4K6av6z1SwTtiYQg,5624
-braindecode/preprocessing/__init__.py,sha256=sSxOr6tW4rUFigFMRjjbgD9c1bQvveIAHRHNisE60s0,740
-braindecode/preprocessing/mne_preprocess.py,sha256=nyrnN_ee-5CkIJWzzyl5aTuFH1rDPIZh5SlIE79mRgo,2258
-braindecode/preprocessing/preprocess.py,sha256=0-08-h8i9u68PIYZ0gxDsLjyCQ2D8UPQGb3HA-eNC_M,16116
-braindecode/preprocessing/windowers.py,sha256=3h8Ne-KtbsTGMnTD3AaoI1izbfDP5MvRqgcuCEZOc4M,27737
-braindecode/samplers/__init__.py,sha256=9ExHqtv8UhvCN-7dy8LhFE-r9AG0x6_c2DIS-eb_kEE,298
-braindecode/samplers/base.py,sha256=eBQtwz_-yxxCQOV7JlURcIjkYlSB-778UV2nGP5zwfY,10268
-braindecode/samplers/ssl.py,sha256=dK0IrSMk01rCqMo7sAjTCP70nApeo2JPBDtV7X8NQJo,4282
-braindecode/training/__init__.py,sha256=MUe_eYFND18KzQYHQxtqvpZ4-gu1CdfvpSsFdFl6nLc,543
-braindecode/training/callbacks.py,sha256=OdOz0d443PdiTuw2LLuPVHHLVYvMj3t7j0lxzmFyLY8,877
-braindecode/training/losses.py,sha256=Pavq9ttQSq2bQXFc7ii5asDepcjiDEEawah36Fgz21E,3779
-braindecode/training/scoring.py,sha256=7oOssI7UVrwzgbW32TDnnl_Ys19dlYBtwN2ipNmk_XQ,19082
-braindecode/visualization/__init__.py,sha256=59XpOfKASsCpd4_oZmk9gvmy5GSy_yhrwG2Ygpb322M,251
-braindecode/visualization/confusion_matrices.py,sha256=eK5UQy26t2fsGxqheEdDjci0zgIm4aWeRdW2TJDSYU0,9543
-braindecode/visualization/gradients.py,sha256=7rzYDzdexJIEk1T7PM7Vse1D53gHnLBkTeWxTqBK5JA,1976
-braindecode-0.8.1.dist-info/LICENSE.txt,sha256=1IX7wR-nRXQap6IOIEXe-ZtJHIaXuiX5UOMsQCrou2I,1520
-braindecode-0.8.1.dist-info/METADATA,sha256=TFtdNiyHenyq2dp1AMwMalbCPHr3Kh0UQf8HxHZDPns,8051
-braindecode-0.8.1.dist-info/WHEEL,sha256=Xo9-1PvkuimrydujYJAjF7pCkriuXBpUPEjma1nZyJ0,92
-braindecode-0.8.1.dist-info/top_level.txt,sha256=pHsWQmSy0uhIez62-HA9j0iaXKvSbUL39ifFRkFnChA,12
-braindecode-0.8.1.dist-info/RECORD,,