Welcome to our Support Center
Updated

BiasSoftmax

Description

Y = softmax(scores + bias)) with simple broadcast on bias. Intended to specialize softmax(scores + additive_mask) commonly found in transformer models.

 

Input parameters

 

specified_outputs_name : array, this parameter lets you manually assign custom names to the output tensors of a node.

 Graphs in : cluster, ONNX model architecture.

data – T : object, the input data as tensor.
bias – T : object, the bias (or mask) as tensor.

 Parameters : cluster,

axis : integer, apply softmax to elements for dimensions axis or higher.
Default value “0”.
is inner broadcast : boolean, true if broadcast bias across input for dimensions broadcast_axis to axis-1, otherwise broadcast bias across input for dimensions 0 to broadcast_axis – 1.
Default value “False”.
 training? : boolean, whether the layer is in training mode (can store data for backward).
Default value “True”.
 lda coeff : float, defines the coefficient by which the loss derivative will be multiplied before being sent to the previous layer (since during the backward run we go backwards).
Default value “1”.

 name (optional) : string, name of the node.

Output parameters

 

 output – T : object, the output.

Type Constraints

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

Example

All these exemples are snippets PNG, you can drop these Snippet onto the block diagram and get the depicted code added to your VI (Do not forget to install Deep Learning library to run it).
Table of Contents