您好，计算lens_logits 时为什么要多加一个1？

WingsBrokenAngel / delving-deeper-into-the-decoder-for-video-captioning

Source code for Delving Deeper into the Decoder for Video Captioning

MIT License

38 stars 14 forks source link

Closed MTCai closed 4 years ago

MTCai commented 4 years ago

lens_logits = tf.nn.softmax( -(tf.abs(lens_reshape - self.options.avglen) + 1), -1)

WingsBrokenAngel commented 4 years ago

To make softmax smoother which was inspired by Laplace smoothing.