Softmax Functions

Machine-readable model

• arm_softmax_f32functionSoftmax using the float-native API signature.
• arm_softmax_f16functionSoftmax using the float-native API signature.

function

arm_softmax_f32

C

Softmax using the float-native API signature.

Include/arm_nnfunctions_flt.h:2439

arm_cmsis_nn_status arm_softmax_f32(const float32_t *input, int32_t num_rows, int32_t row_size, float32_t *output)

Softmax using the float-native API signature.

Each row is normalized independently, and input values never affect the return status at any optimization or floating-point mode. In a row that also holds a finite entry, a -Inf entry gets a small probability that is not guaranteed to be exactly 0 (the scalar exponential clamps its argument). A row holding NaN or +Inf, or only -Inf entries, yields unspecified output values for that row; every other row is still computed.

Parameters

Parameters of arm_softmax_f32

Name	Type	Direction	Description
input	const float32_t *	in	Pointer to the input matrix stored as `num_rows` rows of `row_size` values.
num_rows	int32_t	in	Number of rows in the input matrix.
row_size	int32_t	in	Number of columns per row.
output	float32_t *	out	Pointer to the output matrix.

Returns

Returns of arm_softmax_f32

Description
`ARM_CMSIS_NN_SUCCESS` on success or `ARM_CMSIS_NN_ARG_ERROR` on invalid arguments (NULL pointer, `num_rows` < 1 or `row_size` < 1); on error nothing is written.

function

arm_softmax_f16

C

Softmax using the float-native API signature.

Include/arm_nnfunctions_flt.h:4948

arm_cmsis_nn_status arm_softmax_f16(const float16_t *input, int32_t num_rows, int32_t row_size, float16_t *output)

Softmax using the float-native API signature.

Each row is normalized independently, and input values never affect the return status at any optimization or floating-point mode. In a row that also holds a finite entry, a -Inf entry gets a small probability that is not guaranteed to be exactly 0 (the scalar exponential clamps its argument). A row holding NaN or +Inf, or only -Inf entries, yields unspecified output values for that row; every other row is still computed.

Parameters

Parameters of arm_softmax_f16

Name	Type	Direction	Description
input	const float16_t *	in	Pointer to the input matrix stored as `num_rows` rows of `row_size` values.
num_rows	int32_t	in	Number of rows in the input matrix.
row_size	int32_t	in	Number of columns per row.
output	float16_t *	out	Pointer to the output matrix.

Returns

Returns of arm_softmax_f16

Description
`ARM_CMSIS_NN_SUCCESS` on success or `ARM_CMSIS_NN_ARG_ERROR` on invalid arguments (NULL pointer, `num_rows` < 1 or `row_size` < 1); on error nothing is written.