Softmax - version 11#

This page documents version 11 of operator Softmax. See Softmax for the latest version (since version 13).

  • Domain: ai.onnx

  • Since version: 11

The operator computes the normalized exponential values for the given input. The “axis” attribute indicates the dimension along which Softmax is performed. The output tensor has the same shape as the input tensor.

Inputs

  • input (T): The input tensor that’s coerced into a 2D matrix of size (NxD) as described above.

Outputs

  • output (T): The output values with the same shape as input tensor (the original size without coercion).

Attributes

  • axis (int): Describes the dimension Softmax will be performed on. Negative value means counting dimensions from the back.

Type Constraints

  • T: Constrain input and output types to float tensors. Allowed types: tensor(double), tensor(float), tensor(float16).

Differences with previous version (1)#

SchemaDiff: Softmax (domain 'ai.onnx')

  • old version: 1

  • new version: 11

  • breaking: no

Documentation:

  • line similarity: 0.50 (+2/-2 lines)

--- Softmax v1
+++ Softmax v11
@@ -1,4 +1,4 @@

 The operator computes the normalized exponential values for the given input.
-Inputs are conceptually coerced to a 2D matrix and Softmax is applied on the
-second dimension. The output tensor has the same shape as the input tensor.
+The "axis" attribute indicates the dimension along which Softmax is
+performed. The output tensor has the same shape as the input tensor.