深入浅出Pytorch函数——torch.nn.init.kaiming_normal_

torch.nn.init模块中的所有函数都用于初始化神经网络参数，因此它们都在torc.no_grad()模式下运行，autograd不会将其考虑在内。

根据He, K等人于2015年在《Delving deep into rectifiers: Surpassing human-level performance on ImageNet classification》中描述的方法，用一个正态分布生成值，填充输入的张量或变量。结果张量中的值采样自 N ( 0 , std 2 ) N(0, \text{std}^2) N(0,std2)，其中：
std = gain fan_mode \text{std} = \frac{\text{gain}}{\sqrt{\text{fan\_mode}}} std=fan_mode gain

这种方法也被称为He initialisation。

语法

复制代码

torch.nn.init.kaiming_normal_(tensor, a=0, mode='fan_in', nonlinearity='leaky_relu')

参数

tensor：[Tensor] 一个 N N N维张量torch.Tensor
a：[float] 这层之后使用的rectifier的斜率系数（ReLU的默认值为0）
mode：[str] 可以为fan_in或fan_out。若为fan_in则保留前向传播时权值方差的量级，若为fan_out则保留反向传播时的量级，默认值为fan_in。
nonlinearity：[str] 一个非线性函数，即一个nn.functional的名称，推荐使用relu或者leaky_relu，默认值为leaky_relu。

返回值

一个torch.Tensor且参数tensor也会更新

实例

复制代码

w = torch.empty(3, 5)
nn.init.kaiming_normal_(w, mode='fan_out', nonlinearity='relu')

函数实现

复制代码

def kaiming_normal_(
    tensor: Tensor, a: float = 0, mode: str = 'fan_in', nonlinearity: str = 'leaky_relu'
):
    r"""Fills the input `Tensor` with values according to the method
    described in `Delving deep into rectifiers: Surpassing human-level
    performance on ImageNet classification` - He, K. et al. (2015), using a
    normal distribution. The resulting tensor will have values sampled from
    :math:`\mathcal{N}(0, \text{std}^2)` where

    .. math::
        \text{std} = \frac{\text{gain}}{\sqrt{\text{fan\_mode}}}

    Also known as He initialization.

    Args:
        tensor: an n-dimensional `torch.Tensor`
        a: the negative slope of the rectifier used after this layer (only
            used with ``'leaky_relu'``)
        mode: either ``'fan_in'`` (default) or ``'fan_out'``. Choosing ``'fan_in'``
            preserves the magnitude of the variance of the weights in the
            forward pass. Choosing ``'fan_out'`` preserves the magnitudes in the
            backwards pass.
        nonlinearity: the non-linear function (`nn.functional` name),
            recommended to use only with ``'relu'`` or ``'leaky_relu'`` (default).

    Examples:
        >>> w = torch.empty(3, 5)
        >>> nn.init.kaiming_normal_(w, mode='fan_out', nonlinearity='relu')
    """
    if 0 in tensor.shape:
        warnings.warn("Initializing zero-element tensors is a no-op")
        return tensor
    fan = _calculate_correct_fan(tensor, mode)
    gain = calculate_gain(nonlinearity, a)
    std = gain / math.sqrt(fan)
    with torch.no_grad():
        return tensor.normal_(0, std)