def softmax(logits, axis=None, name=None, dim=None):

This function performs the equivalent of

? ? ? softmax = tf.exp(logits) / tf.reduce_sum(tf.exp(logits), axis)

logits: A non-empty `Tensor`. Must be one of the following types: `half`, `float32`, `float64`.

axis: The dimension softmax would be performed on. The default is -1 which indicates the last dimension.

name: A name for the operation (optional).

dim: Deprecated alias for `axis`.

Returns: A `Tensor`. Has the same type and shape as `logits`.

通過Softmax回歸，將logistic的預測二分類的概率的問題推廣到n分類的概率的問題