tiny_dnn  1.0.0
A header only, dependency-free deep learning framework in C++11
Public Member Functions | List of all members
tiny_dnn::weight_init::lecun Class Reference

Use fan-in(number of input weight for each neuron) for scaling. More...

#include <weight_init.h>

Inheritance diagram for tiny_dnn::weight_init::lecun:
Inheritance graph
[legend]
Collaboration diagram for tiny_dnn::weight_init::lecun:
Collaboration graph
[legend]

Public Member Functions

 lecun (float_t value)
 
void fill (vec_t *weight, serial_size_t fan_in, serial_size_t fan_out) override
 
- Public Member Functions inherited from tiny_dnn::weight_init::scalable
 scalable (float_t value)
 
void scale (float_t value)
 

Additional Inherited Members

- Protected Attributes inherited from tiny_dnn::weight_init::scalable
float_t scale_
 

Detailed Description

Use fan-in(number of input weight for each neuron) for scaling.

Y LeCun, L Bottou, G B Orr, and K Muller, Efficient backprop Neural Networks, Tricks of the Trade, Springer, 1998


The documentation for this class was generated from the following file: