← elementwise_math package

Softmax

Apply the softmax function to the given data array.

The softmax normalizes the data so that the sum of the data along the specified axis is 1.0 and each individual data point is between 0.0 and 1.0, yielding what can be interpreted as a probability distribution. Note that the softmax function is not a smoothed version of the max function, but could be more accurately described as a smoothed version of an argmax function that returns the maximum in a one-hot encoded form (i.e., with a 1 at the maximal index and 0 everywhere else). The function is conventionally used in machine learning to normalize a vector of scores generated by some prediction function (e.g., logits) to a categorical probability distribution. More Info... Version 1.0.0

Ports/Properties

data

Data to process.

verbose name

Data

default value

None

port type

DataPort

value type

AnyNumeric (can be None)

data direction

INOUT

axis

Axis along which the data will have the softmax applied. You can also enter the axis label here as in feature.mylabel.

verbose name

Axis

default value

statistic

port type

ComboPort

value type

str (can be None)

stream

Stream name on which to operate; if unspecified, all streams containing the specified axis will be modified. Wildcards are supported, so mystream* will operate on all streams starting with mystream.

verbose name

Stream

default value

port type

StringPort

value type

str (can be None)

stream_name

verbose name

Stream Name

default value

None

port type

AliasPort

value type

str (can be None)

set_breakpoint

Set a breakpoint on this node. If this is enabled, your debugger (if one is attached) will trigger a breakpoint.

verbose name

Set Breakpoint (Debug Only)

default value

False

port type

BoolPort

value type

bool (can be None)

metadata

User-definable meta-data associated with the node. Usually reserved for technical purposes.

verbose name

Metadata

default value

{}

port type

DictPort

value type

dict (can be None)