Fix Potential NAN bug

[Justobe]

[Potential NAN bug] Loss may become NAN during training

Hello~

Thank you very much for sharing the code!

I try to use my own data set ( with the same shape as mnist) in code. After some iterations, it is found that the training loss becomes NAN. After carefully checking the code, I found that the following code may trigger NAN in loss:

In Tensorflow_gesture/Demo/Mnist.py

cross_entropy = -tf.reduce_sum(y_label * tf.log(y))

If y contains 0 (output of softmax ), the result of tf.log(y) is inf because log(0) is illegal . And this may cause the result of loss to become NAN.

It could be fixed by making the following changes:

cross_entropy = -tf.reduce_sum(y_label * tf.log(y + 1e-7))

or

cross_entropy = -tf.reduce_sum(y_label * tf.log(tf.clip_by_value(y,1e-7,1.0)))

Hope to hear from you ~

Thanks in advance! : )

[Justobe]

@Joooey :)