torch/nn/modules/batchnorm.py

import torch
from .module import Module
from torch.nn.parameter import Parameter
from .. import functional as F


# TODO: check contiguous in THNN
# TODO: use separate backend functions?
class _BatchNorm(Module):

    def __init__(self, num_features, eps=1e-5, momentum=0.1, affine=True):
        super(_BatchNorm, self).__init__()
        self.num_features = num_features
        self.affine = affine
        self.eps = eps
        self.momentum = momentum
        if self.affine:
            self.weight = Parameter(torch.Tensor(num_features))
            self.bias = Parameter(torch.Tensor(num_features))
        else:
            self.register_parameter('weight', None)
            self.register_parameter('bias', None)
        self.register_buffer('running_mean', torch.zeros(num_features))
        self.register_buffer('running_var', torch.ones(num_features))
        self.reset_parameters()

    def reset_parameters(self):
        self.running_mean.zero_()
        self.running_var.fill_(1)
        if self.affine:
            self.weight.data.uniform_()
            self.bias.data.zero_()

    def _check_input_dim(self, input):
        if input.size(1) != self.running_mean.nelement():
            raise ValueError('got {}-feature tensor, expected {}'
                             .format(input.size(1), self.num_features))

    def forward(self, input):
        self._check_input_dim(input)
        return F.batch_norm(
            input, self.running_mean, self.running_var, self.weight, self.bias,
            self.training, self.momentum, self.eps)

    def __repr__(self):
        return ('{name}({num_features}, eps={eps}, momentum={momentum},'
                ' affine={affine})'
                .format(name=self.__class__.__name__, **self.__dict__))


class BatchNorm1d(_BatchNorm):
    r"""Applies Batch Normalization over a 2d or 3d input that is seen as a mini-batch.

    .. math::

        y = \frac{x - mean[x]}{ \sqrt{Var[x] + \epsilon}} * gamma + beta

    The mean and standard-deviation are calculated per-dimension over
    the mini-batches and gamma and beta are learnable parameter vectors
    of size C (where C is the input size).

    During training, this layer keeps a running estimate of its computed mean
    and variance. The running sum is kept with a default momentum of 0.1.

    During evaluation, this running mean/variance is used for normalization.

    Args:
        num_features: num_features from an expected input of size `batch_size x num_features [x width]`
        eps: a value added to the denominator for numerical stability. Default: 1e-5
        momentum: the value used for the running_mean and running_var computation. Default: 0.1
        affine: a boolean value that when set to true, gives the layer learnable affine parameters.

    Shape:
        - Input: :math:`(N, C)` or :math:`(N, C, L)`
        - Output: :math:`(N, C)` or :math:`(N, C, L)` (same shape as input)

    Examples:
        >>> # With Learnable Parameters
        >>> m = nn.BatchNorm1d(100)
        >>> # Without Learnable Parameters
        >>> m = nn.BatchNorm1d(100, affine=False)
        >>> input = autograd.Variable(torch.randn(20, 100))
        >>> output = m(input)
    """

    def _check_input_dim(self, input):
        if input.dim() != 2 and input.dim() != 3:
            raise ValueError('expected 2D or 3D input (got {}D input)'
                             .format(input.dim()))
        super(BatchNorm1d, self)._check_input_dim(input)


class BatchNorm2d(_BatchNorm):
    r"""Applies Batch Normalization over a 4d input that is seen as a mini-batch of 3d inputs

    .. math::

        y = \frac{x - mean[x]}{ \sqrt{Var[x] + \epsilon}} * gamma + beta

    The mean and standard-deviation are calculated per-dimension over
    the mini-batches and gamma and beta are learnable parameter vectors
    of size C (where C is the input size).

    During training, this layer keeps a running estimate of its computed mean
    and variance. The running sum is kept with a default momentum of 0.1.

    During evaluation, this running mean/variance is used for normalization.

    Args:
        num_features: num_features from an expected input of size batch_size x num_features x height x width
        eps: a value added to the denominator for numerical stability. Default: 1e-5
        momentum: the value used for the running_mean and running_var computation. Default: 0.1
        affine: a boolean value that when set to true, gives the layer learnable affine parameters.

    Shape:
        - Input: :math:`(N, C, H, W)`
        - Output: :math:`(N, C, H, W)` (same shape as input)

    Examples:
        >>> # With Learnable Parameters
        >>> m = nn.BatchNorm2d(100)
        >>> # Without Learnable Parameters
        >>> m = nn.BatchNorm2d(100, affine=False)
        >>> input = autograd.Variable(torch.randn(20, 100, 35, 45))
        >>> output = m(input)
    """

    def _check_input_dim(self, input):
        if input.dim() != 4:
            raise ValueError('expected 4D input (got {}D input)'
                             .format(input.dim()))
        super(BatchNorm2d, self)._check_input_dim(input)


class BatchNorm3d(_BatchNorm):
    r"""Applies Batch Normalization over a 5d input that is seen as a mini-batch of 4d inputs

    .. math::

        y = \frac{x - mean[x]}{ \sqrt{Var[x] + \epsilon}} * gamma + beta

    The mean and standard-deviation are calculated per-dimension over
    the mini-batches and gamma and beta are learnable parameter vectors
    of size C (where C is the input size).

    During training, this layer keeps a running estimate of its computed mean
    and variance. The running sum is kept with a default momentum of 0.1.

    During evaluation, this running mean/variance is used for normalization.

    Args:
        num_features: num_features from an expected input of size batch_size x num_features x depth x height x width
        eps: a value added to the denominator for numerical stability. Default: 1e-5
        momentum: the value used for the running_mean and running_var computation. Default: 0.1
        affine: a boolean value that when set to true, gives the layer learnable affine parameters.

    Shape:
        - Input: :math:`(N, C, D, H, W)`
        - Output: :math:`(N, C, D, H, W)` (same shape as input)

    Examples:
        >>> # With Learnable Parameters
        >>> m = nn.BatchNorm3d(100)
        >>> # Without Learnable Parameters
        >>> m = nn.BatchNorm3d(100, affine=False)
        >>> input = autograd.Variable(torch.randn(20, 100, 35, 45, 10))
        >>> output = m(input)
    """

    def _check_input_dim(self, input):
        if input.dim() != 5:
            raise ValueError('expected 5D input (got {}D input)'
                             .format(input.dim()))
        super(BatchNorm3d, self)._check_input_dim(input)