# API

Public kernel functions declared in `arm_nnfunctions*.h`: 426 across 2 headers. Start with an operator family, then follow a function through to its parameters, return values and source.

[Machine-readable model](https://ambiqai.github.io/ns-cmsis-nn/reference/api/reference.json) · [llms.txt](https://ambiqai.github.io/ns-cmsis-nn/reference/api/llms.txt) · [llms-full.txt](https://ambiqai.github.io/ns-cmsis-nn/reference/api/llms-full.txt)

- [Convolution](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#convolution-functions): Conv2D, depthwise, transpose convolution, wrappers, and buffer helpers.
- [Fully connected](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#fully-connected-layer-functions): Dense layers, batch matmul paths, and scratch sizing helpers.
- [Elementwise](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#elementwise-functions): Add, sub, mul, square difference, min/max, batch norm, select, and arithmetic glue.
- [Reduction and comparison](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#basic-math-and-reduction): Argmin/argmax, min/max reductions, comparisons, means, vector sums, and where.
- [Activation](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#activation-functions): ReLU, LeakyReLU, PReLU, Hard-Swish, Logistic, Tanh, and clamp.
- [Data movement](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#data-movement): Pad, reshape, transpose, concatenate, gather, resize, strided slice, tile, broadcast, scatter, mirror pad, and sequence/slice-update utilities.
- [Pooling, softmax, quantization](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#classifier-tail): Classifier tail APIs plus dtype conversion and requantization utilities.
- [Sequence](https://ambiqai.github.io/ns-cmsis-nn/reference/api/#sequence-functions): LSTM, GRU, and SVDF functions for temporal workloads.

## Convolution Functions

Conv2D, depthwise, transpose convolution, wrappers, and buffer helpers. 126 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_convolve_1_x_n_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16) | 1xN convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16_acc16) | 1xN convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16_get_buffer_size) | Get the buffer size required by 1xN convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f32) | 1xN convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f32_get_buffer_size) | Get the buffer size required by 1xN convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_nhwc_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f16) | 1xN convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_nhwc_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f16_acc16) | 1xN convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_nhwc_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f32) | 1xN convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1_x_n_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s4) | 1xn convolution for s4 weights | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1_x_n_s4_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s4_get_buffer_size) | Get the required additional buffer size for 1xn convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1_x_n_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s8) | 1xn convolution | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1_x_n_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s8_get_buffer_size) | Get the required additional buffer size for 1xn convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16) | 1x1 convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16_acc16) | 1x1 convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16_get_buffer_size) | Get the buffer size required by 1x1 convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f32) | 1x1 convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f32_get_buffer_size) | Get the buffer size required by 1x1 convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_nhwc_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f16) | 1x1 convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_nhwc_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f16_acc16) | 1x1 convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_nhwc_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f32) | 1x1 convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_1x1_out_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_out_s8) | Optimised convolution for 1x1 output images (shape of BX1x1xCOUT) for 8x8 computations. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_out_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_out_s8_get_buffer_size) | Get the required scratch buffer size for armconvolve1x1outs8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s16_ns_np_nd`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s16_ns_np_nd) | Pointwise s16 convolution function: no stride, no padding, no dilation. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4) | s4 version for 1x1 convolution with support for non-unity stride values | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s4_fast`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4_fast) | Fast s4 version for 1x1 convolution (non-square shape). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s4_fast_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4_fast_get_buffer_size) | Get the required buffer size for armconvolve1x1s4fast. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8) | s8 version for 1x1 convolution with support for non-unity stride values | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s8_fast`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8_fast) | Fast s8 version for 1x1 convolution (non-square shape). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_1x1_s8_fast_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8_fast_get_buffer_size) | Get the required buffer size for armconvolve1x1s8fast. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_even_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_even_s4) | Basic s4 convolution function with a requirement of even number of kernels. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_even_s4_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_even_s4_get_buffer_size) | Get the required buffer size for armconvolveevens4. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16) | Convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16_acc16) | Convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16_get_buffer_size) | Get the temporary buffer size required by convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f32) | Convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f32_get_buffer_size) | Get the temporary buffer size required by convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_nhwc_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f16) | Convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_nhwc_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f16_acc16) | Convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_nhwc_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f32) | Convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16) | Basic s16 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s16_fast_small_kernel`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_fast_small_kernel) | armconvolves16fastsmallkernel function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_get_buffer_size) | Get the required buffer size for s16 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s16_group_ch_mult_1`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_group_ch_mult_1) | s16 grouped convolution optimized for the case where filterdims->c == 1 and inputch == outputch (channel multiplier = 1). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s4) | Basic s4 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s4_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s4_get_buffer_size) | Get the required buffer size for s4 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8) | Basic s8 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8_3x3_c16_s1`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_3x3_c16_s1) | s8 3x3 convolution over 16 input channels with unit stride. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_buffer_size) | Get the required buffer size for s8 convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_buffer_size_mve) | Get the required buffer size for armconvolves8 for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8_get_weights_sum_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_weights_sum_size) | Get the required buffer size for s8 convolution and depthwise convolution weight sum. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_s8_small_cin`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_small_cin) | s8 convolution for input depths of 1 to 3, such as the first layer of an image or audio model. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_weight_sum`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_weight_sum) | Pre-computes per-output-channel weight sums for a standard convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16) | Convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_wrapper_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16_acc16) | Convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_wrapper_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16_get_buffer_size) | Get the buffer size required by the convolution wrapper. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_wrapper_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f32) | Convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_wrapper_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f32_get_buffer_size) | Get the buffer size required by the convolution wrapper. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_convolve_wrapper_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16) | s16 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size) | Get the required buffer size for armconvolvewrappers16. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s16_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size_dsp) | Get the required buffer size for armconvolvewrappers16 for for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s16_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size_mve) | Get the required buffer size for armconvolvewrappers16 for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4) | s4 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s4_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size) | Get the required buffer size for armconvolvewrappers4. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s4_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size_dsp) | Get the required buffer size for armconvolvewrappers4 for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s4_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size_mve) | Get the required buffer size for armconvolvewrappers4 for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8) | s8 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size) | Get the required buffer size for armconvolvewrappers8. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size_dsp) | Get the required buffer size for armconvolvewrappers8 for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_convolve_wrapper_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size_mve) | Get the required buffer size for armconvolvewrappers8 for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_3x3_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_3x3_s8) | Optimized s8 depthwise convolution function for 3x3 kernel size with some constraints on the input arguments(documented below). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16) | Depthwise convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16_acc16) | Depthwise convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16_get_buffer_size) | Get the temporary buffer size required by depthwise convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f32) | Depthwise convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f32_get_buffer_size) | Get the temporary buffer size required by depthwise convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_fast_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_fast_s16) | Optimized s16 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_fast_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_fast_s16_get_buffer_size) | Get the required buffer size for optimized s16 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s16) | Basic s16 depthwise convolution function that doesn't have any constraints on the input dimensions. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4) | Basic s4 depthwise convolution function that doesn't have any constraints on the input dimensions. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s4_opt`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4_opt) | Optimized s4 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s4_opt_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4_opt_get_buffer_size) | Get the required buffer size for optimized s4 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8) | Basic s8 depthwise convolution function that doesn't have any constraints on the input dimensions. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt) | Optimized s8 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_3x3`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3) | s8 3x3 depthwise convolution that reads the input in place, without the im2col copy of armdepthwiseconvs8opt(), for layers in its gate. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_3x3_c64_s1`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3_c64_s1) | armdepthwiseconvs8opt3x3() specialized for 64 channels and a vertical stride of 1, with fixed channel offsets and edge columns that skip their zero taps. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_3x3_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3_get_buffer_size) | Get the scratch size in bytes of armdepthwiseconvs8opt3x3() and armdepthwiseconvs8opt3x3c64s1(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_channelwise`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_channelwise) | The channel-vectorized path of armdepthwiseconvs8opt() on its own, without the planar attempt. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_get_buffer_size) | Get the required buffer size for optimized s8 depthwise convolution function with constraint that inchannel equals outchannel. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_planar`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_planar) | The planar path of armdepthwiseconvs8opt() on its own: s8 depthwise convolution vectorized across the output pixels of each channel plane. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_s8_opt_planar_supported`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_planar_supported) | Whether armdepthwiseconvs8opt() runs a layer on its planar path, vectorized across the output pixels of each channel plane rather than across channels. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16) | Depthwise convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_wrapper_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16_acc16) | Depthwise convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_wrapper_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16_get_buffer_size) | Get the buffer size required by the depthwise convolution wrapper. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_wrapper_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f32) | Depthwise convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_wrapper_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f32_get_buffer_size) | Get the buffer size required by the depthwise convolution wrapper. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_conv_wrapper_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16) | Wrapper function to pick the right optimized s16 depthwise convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size) | Get size of additional buffer required by armdepthwiseconvwrappers16(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s16_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size_dsp) | Get size of additional buffer required by armdepthwiseconvwrappers16() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s16_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size_mve) | Get size of additional buffer required by armdepthwiseconvwrappers16() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4) | Wrapper function to pick the right optimized s4 depthwise convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s4_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size) | Get size of additional buffer required by armdepthwiseconvwrappers4(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s4_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size_dsp) | Get size of additional buffer required by armdepthwiseconvwrappers4() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s4_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size_mve) | Get size of additional buffer required by armdepthwiseconvwrappers4() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8) | Wrapper function to pick the right optimized s8 depthwise convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size) | Get size of additional buffer required by armdepthwiseconvwrappers8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size_dsp) | Get size of additional buffer required by armdepthwiseconvwrappers8() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_conv_wrapper_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size_mve) | Get size of additional buffer required by armdepthwiseconvwrappers8() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_convolve_weight_sum`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_convolve_weight_sum) | Pre-computes per-channel weight sums for a depthwise convolution. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depthwise_nhwc_conv_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f16) | Depthwise convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_nhwc_conv_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f16_acc16) | Depthwise convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_depthwise_nhwc_conv_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f32) | Depthwise convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16) | Transpose convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16_get_buffer_size) | Get the temporary buffer size required by transpose convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f16_get_reverse_conv_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16_get_reverse_conv_buffer_size) | Get the reverse-convolution workspace size used by transpose convolution helpers. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32) | Transpose convolution, dispatch by layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32_get_buffer_size) | Get the temporary buffer size required by transpose convolution. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_f32_get_reverse_conv_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32_get_reverse_conv_buffer_size) | Get the reverse-convolution workspace size used by transpose convolution helpers. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_nhwc_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_nhwc_f16) | Transpose convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_nhwc_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_nhwc_f32) | Transpose convolution, NHWC layout. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8) | Basic s8 transpose convolution function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_conv_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_buffer_size) | Get the required buffer size for ctx in s8 transpose conv function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_conv_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_buffer_size_mve) | Get size of additional buffer required by armtransposeconvs8() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_conv_s8_get_reverse_conv_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_reverse_conv_buffer_size) | Get the required buffer size for outputctx in s8 transpose conv function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_conv_wrapper_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_wrapper_f16) | Transpose convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_wrapper_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_wrapper_f32) | Transpose convolution wrapper using the CMSIS-NN baseline path. | [Convolution Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnconv/) |
| [`arm_transpose_conv_wrapper_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_wrapper_s8) | Wrapper to select optimal transposed convolution algorithm depending on parameters. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Fully-Connected Layer Functions

Dense layers, batch matmul paths, and scratch sizing helpers. 33 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_batch_matmul_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f16) | Batched matrix multiplication. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_batch_matmul_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f16_get_buffer_size) | Get the temporary buffer size required by batched matrix multiplication. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_batch_matmul_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f32) | Batched matrix multiplication. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_batch_matmul_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f32_get_buffer_size) | Get the temporary buffer size required by batched matrix multiplication. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_batch_matmul_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s16) | Batch matmul function with 16 bit input and output. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_matmul_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8) | Batch matmul function with 8 bit input and output. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_matmul_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size) | Get size of the scratch buffer required by armbatchmatmuls8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_matmul_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size_dsp) | Get size of the scratch buffer required by armbatchmatmuls8() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_matmul_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size_mve) | Get size of the scratch buffer required by armbatchmatmuls8() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16) | Fully connected layer, dispatch by layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16_acc16) | Fully connected layer, dispatch by layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16_get_buffer_size) | Get the temporary buffer size required by the fully connected layer. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f32) | Fully connected layer, dispatch by layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f32_get_buffer_size) | Get the temporary buffer size required by the fully connected layer. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_nhwc_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f16) | Fully connected layer, NHWC layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_nhwc_f16_acc16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f16_acc16) | Fully connected layer, NHWC layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_nhwc_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f32) | Fully connected layer, NHWC layout. | [Fully-connected Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/fc/) |
| [`arm_fully_connected_per_channel_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16) | Basic s16 Fully Connected function using per channel quantization. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_per_channel_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size) | Get size of additional buffer required by armfullyconnectedperchannels16(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_per_channel_s16_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size_dsp) | Get size of additional buffer required by armfullyconnectedperchannels16() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_per_channel_s16_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size_mve) | Get size of additional buffer required by armfullyconnectedperchannels16() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_per_channel_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s8) | Basic s8 Fully Connected function using per channel quantization. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16) | Basic s16 Fully Connected function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size) | Get size of additional buffer required by armfullyconnecteds16(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s16_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size_dsp) | Get size of additional buffer required by armfullyconnecteds16() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s16_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size_mve) | Get size of additional buffer required by armfullyconnecteds16() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s4`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s4) | Basic s4 Fully Connected function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8) | Basic s8 Fully Connected function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size) | Get size of additional buffer required by armfullyconnecteds8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size_dsp) | Get size of additional buffer required by armfullyconnecteds8() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size_mve) | Get size of additional buffer required by armfullyconnecteds8() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_wrapper_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_wrapper_s16) | s16 Fully Connected layer wrapper function | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_fully_connected_wrapper_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_wrapper_s8) | s8 Fully Connected layer wrapper function | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Elementwise Functions

Add, sub, mul, square difference, min/max, batch norm, select, and arithmetic glue. 65 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_abs_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_abs_s16) | s16 elementwise absolute value | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_abs_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_abs_s8) | s8 elementwise absolute value | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_add_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_s16) | s16 elementwise add of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_add_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_s8) | s8 elementwise add of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_add_scalar_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_scalar_s16) | s16 elementwise add of scalar and vector | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_add_scalar_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_scalar_s8) | s8 elementwise add of scalar and vector | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_norm_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_batch_norm_f16) | Apply batch normalization. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_batch_norm_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_batch_norm_f32) | Apply batch normalization. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_elementwise_add_broadcast_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_broadcast_f16) | Elementwise add with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_add_broadcast_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_broadcast_f32) | Elementwise add with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_add_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_f16) | Elementwise add with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_add_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_f32) | Elementwise add with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_add_fp16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_fp16) | Legacy float16 elementwise add with fused clamp, kept only for source compatibility with callers that predate armelementwiseaddf16(). | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_add_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_add_s16) | s16 elementwise add of two vectors | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_add_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_add_s8) | s8 elementwise add of two vectors | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_mul_broadcast_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_broadcast_f16) | Elementwise multiply with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_mul_broadcast_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_broadcast_f32) | Elementwise multiply with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_mul_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_f16) | Elementwise multiply with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_mul_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_f32) | Elementwise multiply with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_mul_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_mul_s16) | s16 elementwise multiplication | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_mul_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_mul_s8) | s8 elementwise multiplication | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_prelu_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_prelu_s16) | Elementwise S16 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_prelu_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_prelu_s8) | Elementwise S8 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_squared_difference_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_squared_difference_f16) | Elementwise squared difference of two float16 vectors. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_squared_difference_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_squared_difference_s16) | s16 elementwise squared difference of two vectors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_squared_difference_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_squared_difference_s8) | s8 elementwise squared difference of two vectors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_sub_broadcast_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_broadcast_f16) | Elementwise subtract with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_sub_broadcast_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_broadcast_f32) | Elementwise subtract with TensorFlow Lite NHWC broadcasting and an output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_sub_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_f16) | Elementwise subtract with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_sub_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_f32) | Elementwise subtract with optional output clamp. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_elementwise_sub_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_sub_s16) | s16 elementwise subtract of two vectors | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_elementwise_sub_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_sub_s8) | s8 elementwise subtract of two vectors | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_maximum_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_maximum_f16) | Elementwise maximum with TensorFlow Lite NHWC broadcasting: each dimension of the two inputs must be equal or 1, and outputdims must be their broadcast shape. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_maximum_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_maximum_f32) | Elementwise maximum with TensorFlow Lite NHWC broadcasting: each dimension of the two inputs must be equal or 1, and outputdims must be their broadcast shape. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_maximum_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_maximum_s16) | s16 elementwise maximum w/ support for broadcasting and scalar inputs. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_maximum_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_maximum_s8) | s8 elementwise maximum w/ support for broadcasting and scalar inputs. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_minimum_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_minimum_f16) | Elementwise minimum with TensorFlow Lite NHWC broadcasting: each dimension of the two inputs must be equal or 1, and outputdims must be their broadcast shape. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_minimum_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_minimum_f32) | Elementwise minimum with TensorFlow Lite NHWC broadcasting: each dimension of the two inputs must be equal or 1, and outputdims must be their broadcast shape. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_minimum_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_minimum_s16) | s16 elementwise minimum w/ support for broadcasting and scalar inputs. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_minimum_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_minimum_s8) | s8 elementwise minimum w/ support for broadcasting and scalar inputs. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mul_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_s16) | s16 elementwise multiplication of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mul_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_s8) | s8 elementwise multiplication of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mul_scalar_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_scalar_s16) | s16 elementwise multiplication of scalar and vector | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mul_scalar_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_scalar_s8) | s8 elementwise multiplication of scalar and vector | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_nn_abs_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_abs_f16) | Elementwise absolute value. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_nn_abs_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_abs_f32) | Elementwise absolute value. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_nn_sqrt_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_sqrt_f16) | Elementwise square root of a float16 tensor. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_nn_sqrt_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_sqrt_f32) | Elementwise square root. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_rsqrt_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_rsqrt_f16) | Elementwise reciprocal square root of a float16 tensor, 1 / sqrt(x). | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_rsqrt_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_rsqrt_f32) | Elementwise reciprocal square root, 1 / sqrt(x). | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_rsqrt_s16_per_op`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_rsqrt_s16_per_op) | INT16 reciprocal square root using a per-operator LUT. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_rsqrt_s16_universal`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_rsqrt_s16_universal) | INT16 reciprocal square root using a shared universal LUT. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_select_v2_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_select_v2_s16) | SELECTV2 with broadcast for int16 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_select_v2_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_select_v2_s8) | SELECTV2 with broadcast for int8 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sqrt_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s16) | s16 elementwise square root using piecewise LUT with linear interpolation | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sqrt_s16_tablefree`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s16_tablefree) | s16 elementwise square root without a lookup table | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sqrt_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s8) | s8 elementwise square root | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_squared_difference_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_s16) | s16 elementwise squared difference of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_squared_difference_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_s8) | s8 elementwise squared difference of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_squared_difference_scalar_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_scalar_s16) | s16 elementwise squared difference of scalar and vector. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_squared_difference_scalar_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_scalar_s8) | s8 elementwise squared difference of scalar and vector. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sub_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_s16) | s16 elementwise subtraction of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sub_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_s8) | s8 elementwise subtraction of two tensors with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sub_scalar_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_scalar_s16) | s16 elementwise subtract of scalar and vector (scalar - vector) | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_sub_scalar_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_scalar_s8) | s8 elementwise subtract of scalar and vector (scalar - vector) | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Basic Math and Reduction

Argmin/argmax, min/max reductions, comparisons, means, vector sums, and where. 40 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_argmax_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmax_f16) | Returns the first maximum's axis-relative INT32 index for a f16 tensor. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_argmax_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmax_f32) | Returns the first maximum's axis-relative INT32 index for a f32 tensor. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_argmax_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmax_s16) | Compute ArgMax indices of an s16 tensor along a specific axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_argmax_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmax_s8) | Compute ArgMax indices of an s8 tensor along a specific axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_argmin_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmin_f16) | Returns the first minimum's axis-relative INT32 index for a f16 tensor. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_argmin_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmin_f32) | Returns the first minimum's axis-relative INT32 index for a f32 tensor. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_argmin_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmin_s16) | Compute ArgMin indices of an s16 tensor along a specific axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_argmin_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmin_s8) | Compute ArgMin indices of an s8 tensor along a specific axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_comparison_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_comparison_s16) | s16 elementwise comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_comparison_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_comparison_s8) | s8 elementwise comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_equal_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_equal_s16) | s16 elementwise equality comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_equal_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_equal_s8) | s8 elementwise equality comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_greater_equal_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_equal_s16) | s16 elementwise greater-or-equal comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_greater_equal_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_equal_s8) | s8 elementwise greater-or-equal comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_greater_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_s16) | s16 elementwise greater-than comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_greater_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_s8) | s8 elementwise greater-than comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_less_equal_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_equal_s16) | s16 elementwise less-or-equal comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_less_equal_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_equal_s8) | s8 elementwise less-or-equal comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_less_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_s16) | s16 elementwise less-than comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_less_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_s8) | s8 elementwise less-than comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mean_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mean_s16) | Computes the mean of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mean_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mean_s8) | Computes the mean of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_nn_mean_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_nn_mean_f16) | Computes the mean of a float16 tensor along the specified axes. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_nn_mean_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_nn_mean_f32) | Computes the mean of a float32 tensor along the specified axes. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_not_equal_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_not_equal_s16) | s16 elementwise inequality comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_not_equal_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_not_equal_s8) | s8 elementwise inequality comparison with support for broadcasting. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reduce_max_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_max_f16) | Reduces a f16 NHWC tensor to its maximum along a binary axis mask. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_reduce_max_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_max_f32) | Reduces a f32 NHWC tensor to its maximum along a binary axis mask. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_reduce_max_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_max_s16) | Computes the max of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reduce_max_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_max_s8) | Computes the max of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reduce_min_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_min_f16) | Reduces a f16 NHWC tensor to its minimum along a binary axis mask. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_reduce_min_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_min_f32) | Reduces a f32 NHWC tensor to its minimum along a binary axis mask. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_reduce_min_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_min_s16) | Computes the min of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reduce_min_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_min_s8) | Computes the min of the input tensor along the specified axis. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reduce_sum_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_sum_f16) | Computes the sum of the input tensor along the specified axes. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_reduce_sum_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_sum_f32) | Computes the sum of the input tensor along the specified axes. | [Reduction Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reduction/) |
| [`arm_vector_sum_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_vector_sum_s8) | Calculate the sum of each row in vectordata, multiply by lhsoffset and optionally add s32 biasdata. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_vector_sum_s8_s64`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_vector_sum_s8_s64) | Calculate the sum of each row in vectordata, multiply by lhsoffset and optionally add s64 biasdata. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_where_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_where_s16) | WHERE operator: return coordinates of non-zero elements in condition (int16). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_where_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_where_s8) | WHERE operator: return coordinates of non-zero elements in condition. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Activation Functions

ReLU, LeakyReLU, PReLU, Hard-Swish, Logistic, Tanh, and clamp. 27 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_clamp_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_clamp_s16) | S16 clamp function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_clamp_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_clamp_s8) | S8 clamp function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_hard_swish_compat_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_compat_s8) | S8 Hard-Swish activation function (compatibility version). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_hard_swish_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_hard_swish_f16) | Hard swish activation for float16 data. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_hard_swish_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_hard_swish_f32) | Hard swish activation for float32 data. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_hard_swish_precise_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_precise_s16) | S16 Hard-Swish activation function (precise version). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_hard_swish_precise_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_precise_s8) | S8 Hard-Swish activation function (precise version). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_leaky_relu_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_leaky_relu_s16) | S16 Leaky ReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_leaky_relu_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_leaky_relu_s8) | S8 Leaky ReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_logistic_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_logistic_s16) | Logistic activation function for s16. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_nn_activation_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_nn_activation_f16) | Elementwise activation. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_nn_activation_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_nn_activation_f32) | Elementwise activation. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_nn_activation_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_nn_activation_s16) | s16 neural network activation function using direct table look-up | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_prelu_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_prelu_f16) | Parametric ReLU for float32 data. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_prelu_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/#arm_prelu_f32) | Parametric ReLU for float32 data. | [Activation Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/acti/) |
| [`arm_prelu_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_s16) | S16 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_prelu_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_s8) | S8 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_prelu_scalar_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_scalar_s16) | Scalar S16 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_prelu_scalar_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_scalar_s8) | Scalar S8 PReLU activation function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_generic_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_generic_s16) | S16 ReLU activation function (generic version) This generic version allows to set custom lower and upper bounds to implement ReLU6, etc. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_generic_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_generic_s8) | S8 ReLU activation function (generic version) This generic version allows to set custom lower and upper bounds to implement ReLU6, etc. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_q15`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_q15) | Q15 RELU function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_q7`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_q7) | Q7 RELU function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_s16) | S16 ReLU activation function lower and upper bounds are quantized representations of 0 and 32767. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_s8) | S8 ReLU activation function lower and upper bounds are quantized representations of 0 and 127. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_relu6_q7`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu6_q7) | Q7 RELU6 function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_tanh_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tanh_s16) | Tanh activation function for s16. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Data Movement

Pad, reshape, transpose, concatenate, gather, resize, strided slice, tile, broadcast, scatter, mirror pad, and sequence/slice-update utilities. 77 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_batch_to_space_nd_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_to_space_nd_s16) | Batch to Space ND function for s16 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_batch_to_space_nd_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_to_space_nd_s8) | Batch to Space ND function for s8 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_broadcast_to_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_broadcast_to_s16) | Broadcast an int16 tensor to a target shape. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_broadcast_to_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_broadcast_to_s8) | Broadcast an int8 tensor to a target shape. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16) | Concatenate float32 tensors of any rank along one axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f16_w`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_w) | Concatenate tensors along the W axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f16_x`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_x) | Concatenate tensors along the X axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f16_y`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_y) | Concatenate tensors along the Y axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f16_z`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_z) | Concatenate tensors along the Z axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32) | Concatenate float32 tensors of any rank along one axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f32_w`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_w) | Concatenate tensors along the W axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f32_x`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_x) | Concatenate tensors along the X axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f32_y`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_y) | Concatenate tensors along the Y axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_f32_z`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_z) | Concatenate tensors along the Z axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_concatenation_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s16) | int16/uint16 concatenation function to be used for concatenating N-tensors along the target axis | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s32) | int32/uint32 concatenation function to be used for concatenating N-tensors along the target axis | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8) | int8/uint8 concatenation function to be used for concatenating N-tensors along the target axis | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s8_w`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_w) | int8/uint8 concatenation function to be used for concatenating N-tensors along the W axis (Batch size) This function should be called for each input tensor to… | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s8_x`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_x) | int8/uint8 concatenation function to be used for concatenating N-tensors along the X axis This function should be called for each input tensor to concatenate. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s8_y`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_y) | int8/uint8 concatenation function to be used for concatenating N-tensors along the Y axis This function should be called for each input tensor to concatenate. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_concatenation_s8_z`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_z) | int8/uint8 concatenation function to be used for concatenating N-tensors along the Z axis This function should be called for each input tensor to concatenate. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depth_to_space_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depth_to_space_s16) | Depth to Space function for s16 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_depth_to_space_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depth_to_space_s8) | Depth to Space function for s8 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_dynamic_update_slice_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dynamic_update_slice_s16) | Update a slice of an int16 operand tensor at runtime-determined indices. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_dynamic_update_slice_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dynamic_update_slice_s8) | Update a slice of an int8 operand tensor at runtime-determined indices. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_gather_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_f16) | Gather contiguous slices along an axis. | [Gather Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/) |
| [`arm_gather_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_f32) | Gather contiguous slices along an axis. | [Gather Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/) |
| [`arm_gather_nd_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_nd_f16) | Gather contiguous slices using coordinate tuples. | [Gather Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/) |
| [`arm_gather_nd_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_nd_f32) | Gather contiguous slices using coordinate tuples. | [Gather Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/gather/) |
| [`arm_gather_nd_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_nd_s16) | Gathernd slices for int16 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_gather_nd_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_nd_s8) | Gathernd slices for int8 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_gather_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_s16) | Gather elements along an axis for int16 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_gather_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_s8) | Gather elements along an axis for int8 tensors. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mirror_pad_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mirror_pad_s16) | Mirror-pad an int16 tensor (REFLECT or SYMMETRIC mode). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_mirror_pad_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mirror_pad_s8) | Mirror-pad an int8 tensor (REFLECT or SYMMETRIC mode). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_nn_fill_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_fill_f16) | Fill a float16 vector with one value; bit copy of value, NaN payload included. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_nn_fill_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_fill_f32) | Fill a float32 vector with one value. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_pack_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_pack_f16) | Stack float32 tensors of equal shape along a new axis (TFLite PACK). | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_pack_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_pack_f32) | Stack float32 tensors of equal shape along a new axis (TFLite PACK). | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_pad_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pad/#arm_pad_f16) | Pad a tensor with a constant value. | [Pad Layer Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pad/) |
| [`arm_pad_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pad/#arm_pad_f32) | Pad a tensor with a constant value. | [Pad Layer Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pad/) |
| [`arm_pad_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_pad_s16) | Expands the size of the input by adding constant values before and after the data, in all dimensions. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_pad_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_pad_s8) | Expands the size of the input by adding constant values before and after the data, in all dimensions. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reshape_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_reshape_f16) | Reshape by copying data without changing element order. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_reshape_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_reshape_f32) | Reshape by copying data without changing element order. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_reshape_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reshape_s8) | Reshape a s8 vector into another with different shape. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_resize_nearest_neighbor_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f16) | Nearest-neighbor resize of a float32 NHWC tensor. | [Reshape Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/) |
| [`arm_resize_nearest_neighbor_f16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f16_get_buffer_size) | Scratch size in bytes for armresizenearestneighborf32() / armresizenearestneighborf16(). | [Reshape Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/) |
| [`arm_resize_nearest_neighbor_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f32) | Nearest-neighbor resize of a float32 NHWC tensor. | [Reshape Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/) |
| [`arm_resize_nearest_neighbor_f32_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f32_get_buffer_size) | Scratch size in bytes for armresizenearestneighborf32() / armresizenearestneighborf16(). | [Reshape Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/reshape/) |
| [`arm_resize_nearest_neighbor_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_resize_nearest_neighbor_s16) | Nearest neighbor resize function for s16 data. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_resize_nearest_neighbor_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_resize_nearest_neighbor_s8) | Nearest neighbor resize function for s8 data. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reverse_sequence_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reverse_sequence_s16) | Reverse variable-length sequences along a dimension for int16. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_reverse_sequence_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reverse_sequence_s8) | Reverse variable-length sequences along a dimension for int8. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_scatter_nd_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_scatter_nd_s16) | Scatter updates into a zero-initialized output tensor for int16. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_scatter_nd_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_scatter_nd_s8) | Scatter updates into a zero-initialized output tensor for int8. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_space_to_batch_nd_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_batch_nd_s16) | Space to Batch ND function for s16 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_space_to_batch_nd_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_batch_nd_s8) | Space to Batch ND function for s8 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_space_to_depth_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_depth_s16) | Space to Depth function for s16 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_space_to_depth_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_depth_s8) | Space to Depth function for s8 data type. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_split_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_split_f16) | Split a float32 tensor of any rank into several tensors along one axis. | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_split_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_split_f32) | Split a float32 tensor of any rank into several tensors along one axis. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_split_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_split_s16) | int16/uint16 split function to be used for splitting a tensor into multiple tensors along the target axis | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_split_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_split_s8) | int8/uint8 split function to be used for splitting a tensor into multiple tensors along the target axis | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_strided_slice_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_strided_slice_f16) | Strided slice for float32 data (pure copy, TensorFlow Lite compatible). | [Elementwise Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/groupelementwise/) |
| [`arm_strided_slice_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/stridedslice/#arm_strided_slice_f32) | Strided slice for float32 data (pure copy, TensorFlow Lite compatible). | [Slicing Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/stridedslice/) |
| [`arm_strided_slice_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s16) | Strided slice function for int16 data. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_strided_slice_s32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s32) | Strided slice function for int32 data. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_strided_slice_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s8) | Strided slice function for int8 data. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_tile_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tile_s16) | Tile an int16 tensor along each dimension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_tile_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tile_s8) | Tile an int8 tensor along each dimension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_transpose_f16) | Transpose a floating-point tensor. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_transpose_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_transpose_f32) | Transpose a floating-point tensor. | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_transpose_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_s16) | Basic s16 transpose function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_transpose_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_s8) | Basic transpose function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_unpack_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_unpack_f16) | Unstack a float32 tensor along one axis into inputshape[axis] tensors (TFLite UNPACK). | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |
| [`arm_unpack_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_unpack_f32) | Unstack a float32 tensor along one axis into inputshape[axis] tensors (TFLite UNPACK). | [NNSupport](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/nnsupport/) |

## Classifier Tail

Classifier tail APIs plus dtype conversion and requantization utilities. 27 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_avg_pool_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_avg_pool_f16) | Average pooling. | [Pooling Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/) |
| [`arm_avg_pool_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_avg_pool_f32) | Average pooling. | [Pooling Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/) |
| [`arm_avgpool_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16) | s16 average pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s16_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size) | Get the required buffer size for S16 average pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s16_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size_dsp) | Get the required buffer size for S16 average pooling function for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s16_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size_mve) | Get the required buffer size for S16 average pooling function for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8) | s8 average pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size) | Get the required buffer size for S8 average pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size_dsp) | Get the required buffer size for S8 average pooling function for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_avgpool_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size_mve) | Get the required buffer size for S8 average pooling function for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_dequantize_f16_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/quantization/#arm_dequantize_f16_f32) | Widen a float16 vector to float32. | [Quantization Functions:](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/quantization/) |
| [`arm_dequantize_s16_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dequantize_s16_f32) | Dequantize an int16t array back to floating-point format. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_dequantize_s8_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dequantize_s8_f32) | Dequantize an int8t array back to floating-point format. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_max_pool_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_max_pool_f16) | Max pooling. | [Pooling Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/) |
| [`arm_max_pool_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_max_pool_f32) | Max pooling. | [Pooling Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/pooling/) |
| [`arm_max_pool_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_max_pool_s16) | s16 max pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_max_pool_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_max_pool_s8) | s8 max pooling function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_quantize_f32_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_quantize_f32_s16) | Quantize a floating-point array into int16t format. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_quantize_f32_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_quantize_f32_s8) | Quantize a floating-point array into int8t format. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_requantize_s16_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_requantize_s16_s16) | Requantize an int16t array to another int16t range with a different scale. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_requantize_s8_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_requantize_s8_s8) | Requantize an int8t array to another int8t range with a different scale. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_softmax_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/softmax/#arm_softmax_f16) | Softmax using the float-native API signature. | [Softmax Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/softmax/) |
| [`arm_softmax_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/softmax/#arm_softmax_f32) | Softmax using the float-native API signature. | [Softmax Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/softmax/) |
| [`arm_softmax_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s16) | S16 softmax function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_softmax_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s8) | S8 softmax function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_softmax_s8_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s8_s16) | S8 to s16 softmax function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_softmax_u8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_u8) | U8 softmax function. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Sequence Functions

LSTM, GRU, and SVDF functions for temporal workloads. 31 functions.

| Function | Summary | Module |
| --- | --- | --- |
| [`arm_gru_unidirectional_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f16) | Unidirectional GRU layer for float16 input, output and state. | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_gru_unidirectional_f16_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f16_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armgruunidirectionalf16(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_gru_unidirectional_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f32) | Unidirectional GRU layer for float32 input, output and state. | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_gru_unidirectional_f32_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f32_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armgruunidirectionalf32(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16) | Unidirectional LSTM inference. | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f16_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armlstmunidirectionalf16(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f16_temp2_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16_temp2_get_buffer_size) | Get size of the temp2 scratch buffer required by armlstmunidirectionalf16(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32) | Unidirectional LSTM inference. | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f32_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armlstmunidirectionalf32(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_f32_temp2_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32_temp2_get_buffer_size) | Get size of the temp2 scratch buffer required by armlstmunidirectionalf32(). | [LSTM Layer Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/lstm/) |
| [`arm_lstm_unidirectional_s16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16) | LSTM unidirectional function with 16 bit input and output and 16 bit gate output, 64 bit bias. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_lstm_unidirectional_s16_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armlstmunidirectionals16(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_lstm_unidirectional_s16_temp2_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16_temp2_get_buffer_size) | Get size of the temp2 scratch buffer required by armlstmunidirectionals16(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_lstm_unidirectional_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8) | LSTM unidirectional function with 8 bit input and output and 16 bit gate output, 32 bit bias. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_lstm_unidirectional_s8_temp1_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8_temp1_get_buffer_size) | Get size of the temp1 scratch buffer required by armlstmunidirectionals8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_lstm_unidirectional_s8_temp2_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8_temp2_get_buffer_size) | Get size of the temp2 scratch buffer required by armlstmunidirectionals8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_f16`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16) | Stateful singular value decomposition filter, float16 variant. | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_f16_input_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16_input_ctx_get_buffer_size) | Get size of the inputctx staging buffer required by armsvdff16(). | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_f16_output_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16_output_ctx_get_buffer_size) | Get size of the outputctx staging buffer required by armsvdff16(). | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_f32`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32) | Stateful singular value decomposition filter. | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_f32_input_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32_input_ctx_get_buffer_size) | Get size of the inputctx staging buffer required by armsvdff32(). | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_f32_output_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32_output_ctx_get_buffer_size) | Get size of the outputctx staging buffer required by armsvdff32(). | [SVDF Functions](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/svdf/) |
| [`arm_svdf_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8) | s8 SVDF function with 8 bit state tensor and 8 bit time weights | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_s8_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size) | Get size of the kernel-sum buffer required by armsvdfs8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_s8_get_buffer_size_dsp`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size_dsp) | Get size of the kernel-sum buffer required by armsvdfs8() for processors with DSP extension. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_s8_get_buffer_size_mve`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size_mve) | Get size of the kernel-sum buffer required by armsvdfs8() for Arm(R) Helium Architecture case. | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_s8_input_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_input_ctx_get_buffer_size) | Get size of the inputctx staging buffer required by armsvdfs8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_s8_output_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_output_ctx_get_buffer_size) | Get size of the outputctx staging buffer required by armsvdfs8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_state_s16_s8`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8) | s8 SVDF function with 16 bit state tensor and 16 bit time weights | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_state_s16_s8_input_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8_input_ctx_get_buffer_size) | Get size of the inputctx staging buffer required by armsvdfstates16s8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |
| [`arm_svdf_state_s16_s8_output_ctx_get_buffer_size`](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8_output_ctx_get_buffer_size) | Get size of the outputctx staging buffer required by armsvdfstates16s8(). | [arm_nnfunctions.h](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/) |

## Types and support

Structs, enums, macros and typedefs stay on the module pages rather than being repeated here. Open [heliaCORE](https://ambiqai.github.io/ns-cmsis-nn/reference/api/heliacore/) for the full tree.