Tensorflow.js tf.softmax() Function
The tf.softmax() function is used to compute the softmax normalized vector given the logits.
tf.softmax (logits, dim?)
Parameters: This function accept two parameters which are illustrated below:
- Logits: the logits array.
- dim: The dimension softmax would be performed on.
Return Value: It returns tf.Tensor
Tensor [0.4683105, 0.0633789, 0.4683105]
Tensor [[0.9525742, 0.0474259], [0.7310586, 0.2689414], [0.9933072, 0.0066929]]