LRN

tensorflow C++ API

tensorflow::ops::LRN

Local Response Normalization.

Summary

The 4-Dinputtensor is treated as a 3-D array of 1-D vectors (along the last dimension), and each vector is normalized independently. Within a given vector, each component is divided by the weighted, squared sum of inputs withindepth_radius. In detail,

sqr_sum[a, b, c, d] =
    sum(input[a, b, c, d - depth_radius : d + depth_radius + 1] ** 2)
output = input / (bias + alpha * sqr_sum) ** beta

For details, seeKrizhevsky et al., ImageNet classification with deep convolutional neural networks (NIPS 2012).

Arguments:

• scope: A Scope object
• input: 4-D.

Optional attributes (seeAttrs):

• depth_radius: 0-D. Half-width of the 1-D normalization window.
• bias: An offset (usually positive to avoid dividing by 0).
• alpha: A scale factor, usually positive.
• beta: An exponent.

Returns:

• Output: The output tensor.

LRN block

Source link : https://github.com/EXPNUNI/enuSpaceTensorflow/blob/master/enuSpaceTensorflow/tf_nn.cpp

Argument:

• Scope scope : A Scope object (A scope is generated automatically each page. A scope is not connected.)
• Input input: connect Input node.
• LRN::Attrs attrs :inputs attrs in values ex)depth_radius_ = 5;bias_ = 1.0f;alpha_ = 1.0f;beta_ = 0.5f;

Return:

• Output output: Output object of LRN class object.

Result:

• std::vector(Tensor) result_output : Returned object of executed result by calling session.

Using Method