Skip to content

Latest commit

 

History

History
419 lines (337 loc) · 8.05 KB

nn.aliases.md

File metadata and controls

419 lines (337 loc) · 8.05 KB
.. role:: hidden
    :class: hidden-section

Aliases in torch.nn

.. automodule:: torch.nn.modules

The following are aliases to their counterparts in torch.nn in nested namespaces.

torch.nn.modules

The following are aliases to their counterparts in torch.nn in the torch.nn.modules namespace.

Containers (Aliases)

.. currentmodule:: torch.nn.modules

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    container.Sequential
    container.ModuleList
    container.ModuleDict
    container.ParameterList
    container.ParameterDict

Convolution Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    conv.Conv1d
    conv.Conv2d
    conv.Conv3d
    conv.ConvTranspose1d
    conv.ConvTranspose2d
    conv.ConvTranspose3d
    conv.LazyConv1d
    conv.LazyConv2d
    conv.LazyConv3d
    conv.LazyConvTranspose1d
    conv.LazyConvTranspose2d
    conv.LazyConvTranspose3d
    fold.Unfold
    fold.Fold

Pooling layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    pooling.MaxPool1d
    pooling.MaxPool2d
    pooling.MaxPool3d
    pooling.MaxUnpool1d
    pooling.MaxUnpool2d
    pooling.MaxUnpool3d
    pooling.AvgPool1d
    pooling.AvgPool2d
    pooling.AvgPool3d
    pooling.FractionalMaxPool2d
    pooling.FractionalMaxPool3d
    pooling.LPPool1d
    pooling.LPPool2d
    pooling.LPPool3d
    pooling.AdaptiveMaxPool1d
    pooling.AdaptiveMaxPool2d
    pooling.AdaptiveMaxPool3d
    pooling.AdaptiveAvgPool1d
    pooling.AdaptiveAvgPool2d
    pooling.AdaptiveAvgPool3d

Padding Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    padding.ReflectionPad1d
    padding.ReflectionPad2d
    padding.ReflectionPad3d
    padding.ReplicationPad1d
    padding.ReplicationPad2d
    padding.ReplicationPad3d
    padding.ZeroPad1d
    padding.ZeroPad2d
    padding.ZeroPad3d
    padding.ConstantPad1d
    padding.ConstantPad2d
    padding.ConstantPad3d
    padding.CircularPad1d
    padding.CircularPad2d
    padding.CircularPad3d

Non-linear Activations (weighted sum, nonlinearity) (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    activation.ELU
    activation.Hardshrink
    activation.Hardsigmoid
    activation.Hardtanh
    activation.Hardswish
    activation.LeakyReLU
    activation.LogSigmoid
    activation.MultiheadAttention
    activation.PReLU
    activation.ReLU
    activation.ReLU6
    activation.RReLU
    activation.SELU
    activation.CELU
    activation.GELU
    activation.Sigmoid
    activation.SiLU
    activation.Mish
    activation.Softplus
    activation.Softshrink
    activation.Softsign
    activation.Tanh
    activation.Tanhshrink
    activation.Threshold
    activation.GLU

Non-linear Activations (other) (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    activation.Softmin
    activation.Softmax
    activation.Softmax2d
    activation.LogSoftmax
    adaptive.AdaptiveLogSoftmaxWithLoss

Normalization Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    batchnorm.BatchNorm1d
    batchnorm.BatchNorm2d
    batchnorm.BatchNorm3d
    batchnorm.LazyBatchNorm1d
    batchnorm.LazyBatchNorm2d
    batchnorm.LazyBatchNorm3d
    normalization.GroupNorm
    batchnorm.SyncBatchNorm
    instancenorm.InstanceNorm1d
    instancenorm.InstanceNorm2d
    instancenorm.InstanceNorm3d
    instancenorm.LazyInstanceNorm1d
    instancenorm.LazyInstanceNorm2d
    instancenorm.LazyInstanceNorm3d
    normalization.LayerNorm
    normalization.LocalResponseNorm
    normalization.RMSNorm

Recurrent Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    rnn.RNNBase
    rnn.RNN
    rnn.LSTM
    rnn.GRU
    rnn.RNNCell
    rnn.LSTMCell
    rnn.GRUCell

Transformer Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    transformer.Transformer
    transformer.TransformerEncoder
    transformer.TransformerDecoder
    transformer.TransformerEncoderLayer
    transformer.TransformerDecoderLayer

Linear Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    linear.Identity
    linear.Linear
    linear.Bilinear
    linear.LazyLinear

Dropout Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    dropout.Dropout
    dropout.Dropout1d
    dropout.Dropout2d
    dropout.Dropout3d
    dropout.AlphaDropout
    dropout.FeatureAlphaDropout

Sparse Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    sparse.Embedding
    sparse.EmbeddingBag

Distance Functions (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    distance.CosineSimilarity
    distance.PairwiseDistance

Loss Functions (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    loss.L1Loss
    loss.MSELoss
    loss.CrossEntropyLoss
    loss.CTCLoss
    loss.NLLLoss
    loss.PoissonNLLLoss
    loss.GaussianNLLLoss
    loss.KLDivLoss
    loss.BCELoss
    loss.BCEWithLogitsLoss
    loss.MarginRankingLoss
    loss.HingeEmbeddingLoss
    loss.MultiLabelMarginLoss
    loss.HuberLoss
    loss.SmoothL1Loss
    loss.SoftMarginLoss
    loss.MultiLabelSoftMarginLoss
    loss.CosineEmbeddingLoss
    loss.MultiMarginLoss
    loss.TripletMarginLoss
    loss.TripletMarginWithDistanceLoss

Vision Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    pixelshuffle.PixelShuffle
    pixelshuffle.PixelUnshuffle
    upsampling.Upsample
    upsampling.UpsamplingNearest2d
    upsampling.UpsamplingBilinear2d

Shuffle Layers (Aliases)

.. autosummary::
    :toctree: generated
    :nosignatures:
    :template: classtemplate.rst

    channelshuffle.ChannelShuffle

torch.nn.utils

The following are aliases to their counterparts in torch.nn.utils in nested namespaces.

Utility functions to clip parameter gradients.

.. currentmodule:: torch.nn.utils

.. autosummary::
    :toctree: generated
    :nosignatures:

    clip_grad.clip_grad_norm_
    clip_grad.clip_grad_norm
    clip_grad.clip_grad_value_

Utility functions to flatten and unflatten Module parameters to and from a single vector.

.. autosummary::
    :toctree: generated
    :nosignatures:

    convert_parameters.parameters_to_vector
    convert_parameters.vector_to_parameters

Utility functions to fuse Modules with BatchNorm modules.

.. autosummary::
    :toctree: generated
    :nosignatures:

    fusion.fuse_conv_bn_eval
    fusion.fuse_conv_bn_weights
    fusion.fuse_linear_bn_eval
    fusion.fuse_linear_bn_weights

Utility functions to convert Module parameter memory formats.

.. autosummary::
    :toctree: generated
    :nosignatures:

    memory_format.convert_conv2d_weight_memory_format
    memory_format.convert_conv3d_weight_memory_format

Utility functions to apply and remove weight normalization from Module parameters.

.. autosummary::
    :toctree: generated
    :nosignatures:

    weight_norm.weight_norm
    weight_norm.remove_weight_norm
    spectral_norm.spectral_norm
    spectral_norm.remove_spectral_norm

Utility functions for initializing Module parameters.

.. autosummary::
    :toctree: generated
    :nosignatures:

    init.skip_init