# Kernel index

Search by function name or description, then narrow the results by operator group
and data type. Select a function to read its parameters and requirements.

## Functions

- [arm_abs_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_abs_s16): Elementwise — s16 elementwise absolute value — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_abs_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_abs_s8): Elementwise — s8 elementwise absolute value — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_add_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_s16): Elementwise — s16 elementwise add of two tensors with support for broadcasting. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_add_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_s8): Elementwise — s8 elementwise add of two tensors with support for broadcasting. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_add_scalar_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_scalar_s16): Elementwise — s16 elementwise add of scalar and vector — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_add_scalar_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_add_scalar_s8): Elementwise — s8 elementwise add of scalar and vector — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_argmax_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmax_f16): Reduction and comparison — Returns the first maximum's axis-relative INT32 index for a f16 tensor. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_argmax_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmax_f32): Reduction and comparison — Returns the first maximum's axis-relative INT32 index for a f32 tensor. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_argmax_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmax_s16): Reduction and comparison — Compute ArgMax indices of an s16 tensor along a specific axis. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_argmax_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmax_s8): Reduction and comparison — Compute ArgMax indices of an s8 tensor along a specific axis. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_argmin_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmin_f16): Reduction and comparison — Returns the first minimum's axis-relative INT32 index for a f16 tensor. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_argmin_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_argmin_f32): Reduction and comparison — Returns the first minimum's axis-relative INT32 index for a f32 tensor. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_argmin_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmin_s16): Reduction and comparison — Compute ArgMin indices of an s16 tensor along a specific axis. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_argmin_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_argmin_s8): Reduction and comparison — Compute ArgMin indices of an s8 tensor along a specific axis. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_avg_pool_f16](/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_avg_pool_f16): Pooling, softmax, quantization — Average pooling. — group: Pooling, softmax, quantization — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_avg_pool_f32](/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_avg_pool_f32): Pooling, softmax, quantization — Average pooling. — group: Pooling, softmax, quantization — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_avgpool_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16): Pooling, softmax, quantization — s16 average pooling function. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_avgpool_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size): Pooling, softmax, quantization — Get the required buffer size for S16 average pooling function. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_avgpool_s16_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size_dsp): Pooling, softmax, quantization — Get the required buffer size for S16 average pooling function for processors with DSP extension. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_avgpool_s16_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s16_get_buffer_size_mve): Pooling, softmax, quantization — Get the required buffer size for S16 average pooling function for Arm(R) Helium Architecture case. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_avgpool_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8): Pooling, softmax, quantization — s8 average pooling function. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_avgpool_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size): Pooling, softmax, quantization — Get the required buffer size for S8 average pooling function. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_avgpool_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size_dsp): Pooling, softmax, quantization — Get the required buffer size for S8 average pooling function for processors with DSP extension. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_avgpool_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_avgpool_s8_get_buffer_size_mve): Pooling, softmax, quantization — Get the required buffer size for S8 average pooling function for Arm(R) Helium Architecture case. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_batch_matmul_f16](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f16): Fully connected — Batched matrix multiplication. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_batch_matmul_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f16_get_buffer_size): Fully connected — Get the temporary buffer size required by batched matrix multiplication. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_batch_matmul_f32](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f32): Fully connected — Batched matrix multiplication. — group: Fully connected — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_batch_matmul_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_batch_matmul_f32_get_buffer_size): Fully connected — Get the temporary buffer size required by batched matrix multiplication. — group: Fully connected — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_batch_matmul_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s16): Fully connected — Batch matmul function with 16 bit input and output. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_batch_matmul_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8): Fully connected — Batch matmul function with 8 bit input and output. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_batch_matmul_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size): Fully connected — Get size of the scratch buffer required by armbatchmatmuls8(). — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_batch_matmul_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size_dsp): Fully connected — Get size of the scratch buffer required by armbatchmatmuls8() for processors with DSP extension. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_batch_matmul_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_matmul_s8_get_buffer_size_mve): Fully connected — Get size of the scratch buffer required by armbatchmatmuls8() for Arm(R) Helium Architecture case. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_batch_norm_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_batch_norm_f16): Elementwise — Apply batch normalization. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_batch_norm_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_batch_norm_f32): Elementwise — Apply batch normalization. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_batch_to_space_nd_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_to_space_nd_s16): Data movement — Batch to Space ND function for s16 data type. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_batch_to_space_nd_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_batch_to_space_nd_s8): Data movement — Batch to Space ND function for s8 data type. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_broadcast_to_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_broadcast_to_s16): Data movement — Broadcast an int16 tensor to a target shape. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_broadcast_to_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_broadcast_to_s8): Data movement — Broadcast an int8 tensor to a target shape. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_clamp_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_clamp_s16): Activation — S16 clamp function. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_clamp_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_clamp_s8): Activation — S8 clamp function. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_comparison_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_comparison_s16): Reduction and comparison — s16 elementwise comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_comparison_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_comparison_s8): Reduction and comparison — s8 elementwise comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_concatenation_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16): Data movement — Concatenate float32 tensors of any rank along one axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f16_w](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_w): Data movement — Concatenate tensors along the W axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f16_x](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_x): Data movement — Concatenate tensors along the X axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f16_y](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_y): Data movement — Concatenate tensors along the Y axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f16_z](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f16_z): Data movement — Concatenate tensors along the Z axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32): Data movement — Concatenate float32 tensors of any rank along one axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f32_w](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_w): Data movement — Concatenate tensors along the W axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f32_x](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_x): Data movement — Concatenate tensors along the X axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f32_y](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_y): Data movement — Concatenate tensors along the Y axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_f32_z](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_concatenation_f32_z): Data movement — Concatenate tensors along the Z axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_concatenation_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s16): Data movement — int16/uint16 concatenation function to be used for concatenating N-tensors along the target axis — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_concatenation_s32](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s32): Data movement — int32/uint32 concatenation function to be used for concatenating N-tensors along the target axis — group: Data movement — dtypes: s32 — headers: arm_nnfunctions.h
- [arm_concatenation_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8): Data movement — int8/uint8 concatenation function to be used for concatenating N-tensors along the target axis — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_concatenation_s8_w](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_w): Data movement — 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… — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_concatenation_s8_x](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_x): Data movement — 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. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_concatenation_s8_y](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_y): Data movement — 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. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_concatenation_s8_z](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_concatenation_s8_z): Data movement — 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. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1_x_n_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16): Convolution — 1xN convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16_acc16): Convolution — 1xN convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f16_get_buffer_size): Convolution — Get the buffer size required by 1xN convolution. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f32): Convolution — 1xN convolution, dispatch by layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_f32_get_buffer_size): Convolution — Get the buffer size required by 1xN convolution. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_nhwc_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f16): Convolution — 1xN convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_nhwc_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f16_acc16): Convolution — 1xN convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_nhwc_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1_x_n_nhwc_f32): Convolution — 1xN convolution, NHWC layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1_x_n_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s4): Convolution — 1xn convolution for s4 weights — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_1_x_n_s4_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s4_get_buffer_size): Convolution — Get the required additional buffer size for 1xn convolution. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_1_x_n_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s8): Convolution — 1xn convolution — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1_x_n_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1_x_n_s8_get_buffer_size): Convolution — Get the required additional buffer size for 1xn convolution. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16): Convolution — 1x1 convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16_acc16): Convolution — 1x1 convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f16_get_buffer_size): Convolution — Get the buffer size required by 1x1 convolution. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f32): Convolution — 1x1 convolution, dispatch by layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_f32_get_buffer_size): Convolution — Get the buffer size required by 1x1 convolution. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_nhwc_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f16): Convolution — 1x1 convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_nhwc_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f16_acc16): Convolution — 1x1 convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_nhwc_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_1x1_nhwc_f32): Convolution — 1x1 convolution, NHWC layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_1x1_out_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_out_s8): Convolution — Optimised convolution for 1x1 output images (shape of BX1x1xCOUT) for 8x8 computations. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_out_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_out_s8_get_buffer_size): Convolution — Get the required scratch buffer size for armconvolve1x1outs8(). — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s16_ns_np_nd](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s16_ns_np_nd): Convolution — Pointwise s16 convolution function: no stride, no padding, no dilation. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4): Convolution — s4 version for 1x1 convolution with support for non-unity stride values — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s4_fast](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4_fast): Convolution — Fast s4 version for 1x1 convolution (non-square shape). — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s4_fast_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s4_fast_get_buffer_size): Convolution — Get the required buffer size for armconvolve1x1s4fast. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8): Convolution — s8 version for 1x1 convolution with support for non-unity stride values — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s8_fast](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8_fast): Convolution — Fast s8 version for 1x1 convolution (non-square shape). — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_1x1_s8_fast_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_1x1_s8_fast_get_buffer_size): Convolution — Get the required buffer size for armconvolve1x1s8fast. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_even_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_even_s4): Convolution — Basic s4 convolution function with a requirement of even number of kernels. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_even_s4_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_even_s4_get_buffer_size): Convolution — Get the required buffer size for armconvolveevens4. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16): Convolution — Convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16_acc16): Convolution — Convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f16_get_buffer_size): Convolution — Get the temporary buffer size required by convolution. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f32): Convolution — Convolution, dispatch by layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_f32_get_buffer_size): Convolution — Get the temporary buffer size required by convolution. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_nhwc_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f16): Convolution — Convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_nhwc_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f16_acc16): Convolution — Convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_nhwc_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_nhwc_f32): Convolution — Convolution, NHWC layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16): Convolution — Basic s16 convolution function. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_s16_fast_small_kernel](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_fast_small_kernel): Convolution — armconvolves16fastsmallkernel function. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_get_buffer_size): Convolution — Get the required buffer size for s16 convolution function. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_s16_group_ch_mult_1](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s16_group_ch_mult_1): Convolution — s16 grouped convolution optimized for the case where filterdims->c == 1 and inputch == outputch (channel multiplier = 1). — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s4): Convolution — Basic s4 convolution function. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_s4_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s4_get_buffer_size): Convolution — Get the required buffer size for s4 convolution function. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8): Convolution — Basic s8 convolution function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_s8_3x3_c16_s1](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_3x3_c16_s1): Convolution — s8 3x3 convolution over 16 input channels with unit stride. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_buffer_size): Convolution — Get the required buffer size for s8 convolution function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_buffer_size_mve): Convolution — Get the required buffer size for armconvolves8 for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_s8_get_weights_sum_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_get_weights_sum_size): Convolution — Get the required buffer size for s8 convolution and depthwise convolution weight sum. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_s8_small_cin](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_s8_small_cin): Convolution — s8 convolution for input depths of 1 to 3, such as the first layer of an image or audio model. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_weight_sum](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_weight_sum): Convolution — Pre-computes per-output-channel weight sums for a standard convolution. — group: Convolution — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16): Convolution — Convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_wrapper_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16_acc16): Convolution — Convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_wrapper_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f16_get_buffer_size): Convolution — Get the buffer size required by the convolution wrapper. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_convolve_wrapper_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f32): Convolution — Convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_wrapper_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_convolve_wrapper_f32_get_buffer_size): Convolution — Get the buffer size required by the convolution wrapper. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_convolve_wrapper_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16): Convolution — s16 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size): Convolution — Get the required buffer size for armconvolvewrappers16. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s16_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size_dsp): Convolution — Get the required buffer size for armconvolvewrappers16 for for processors with DSP extension. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s16_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s16_get_buffer_size_mve): Convolution — Get the required buffer size for armconvolvewrappers16 for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4): Convolution — s4 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s4_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size): Convolution — Get the required buffer size for armconvolvewrappers4. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s4_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size_dsp): Convolution — Get the required buffer size for armconvolvewrappers4 for processors with DSP extension. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s4_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s4_get_buffer_size_mve): Convolution — Get the required buffer size for armconvolvewrappers4 for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8): Convolution — s8 convolution layer wrapper function with the main purpose to call the optimal kernel available in cmsis-nn to perform the convolution. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size): Convolution — Get the required buffer size for armconvolvewrappers8. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size_dsp): Convolution — Get the required buffer size for armconvolvewrappers8 for processors with DSP extension. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_convolve_wrapper_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_convolve_wrapper_s8_get_buffer_size_mve): Convolution — Get the required buffer size for armconvolvewrappers8 for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depth_to_space_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depth_to_space_s16): Data movement — Depth to Space function for s16 data type. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depth_to_space_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depth_to_space_s8): Data movement — Depth to Space function for s8 data type. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_3x3_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_3x3_s8): Convolution — Optimized s8 depthwise convolution function for 3x3 kernel size with some constraints on the input arguments(documented below). — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16): Convolution — Depthwise convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16_acc16): Convolution — Depthwise convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f16_get_buffer_size): Convolution — Get the temporary buffer size required by depthwise convolution. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f32): Convolution — Depthwise convolution, dispatch by layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_f32_get_buffer_size): Convolution — Get the temporary buffer size required by depthwise convolution. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_fast_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_fast_s16): Convolution — Optimized s16 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_fast_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_fast_s16_get_buffer_size): Convolution — Get the required buffer size for optimized s16 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s16): Convolution — Basic s16 depthwise convolution function that doesn't have any constraints on the input dimensions. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4): Convolution — Basic s4 depthwise convolution function that doesn't have any constraints on the input dimensions. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s4_opt](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4_opt): Convolution — Optimized s4 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s4_opt_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s4_opt_get_buffer_size): Convolution — Get the required buffer size for optimized s4 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8): Convolution — Basic s8 depthwise convolution function that doesn't have any constraints on the input dimensions. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt): Convolution — Optimized s8 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_3x3](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3): Convolution — s8 3x3 depthwise convolution that reads the input in place, without the im2col copy of armdepthwiseconvs8opt(), for layers in its gate. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_3x3_c64_s1](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3_c64_s1): Convolution — armdepthwiseconvs8opt3x3() specialized for 64 channels and a vertical stride of 1, with fixed channel offsets and edge columns that skip their zero taps. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_3x3_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_3x3_get_buffer_size): Convolution — Get the scratch size in bytes of armdepthwiseconvs8opt3x3() and armdepthwiseconvs8opt3x3c64s1(). — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_channelwise](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_channelwise): Convolution — The channel-vectorized path of armdepthwiseconvs8opt() on its own, without the planar attempt. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_get_buffer_size): Convolution — Get the required buffer size for optimized s8 depthwise convolution function with constraint that inchannel equals outchannel. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_planar](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_planar): Convolution — The planar path of armdepthwiseconvs8opt() on its own: s8 depthwise convolution vectorized across the output pixels of each channel plane. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_s8_opt_planar_supported](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_s8_opt_planar_supported): Convolution — Whether armdepthwiseconvs8opt() runs a layer on its planar path, vectorized across the output pixels of each channel plane rather than across channels. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16): Convolution — Depthwise convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_wrapper_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16_acc16): Convolution — Depthwise convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_wrapper_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f16_get_buffer_size): Convolution — Get the buffer size required by the depthwise convolution wrapper. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_wrapper_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f32): Convolution — Depthwise convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_wrapper_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_conv_wrapper_f32_get_buffer_size): Convolution — Get the buffer size required by the depthwise convolution wrapper. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_conv_wrapper_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16): Convolution — Wrapper function to pick the right optimized s16 depthwise convolution function. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers16(). — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s16_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size_dsp): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers16() for processors with DSP extension. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s16_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s16_get_buffer_size_mve): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers16() for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4): Convolution — Wrapper function to pick the right optimized s4 depthwise convolution function. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s4_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers4(). — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s4_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size_dsp): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers4() for processors with DSP extension. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s4_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s4_get_buffer_size_mve): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers4() for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8): Convolution — Wrapper function to pick the right optimized s8 depthwise convolution function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers8(). — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size_dsp): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers8() for processors with DSP extension. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_conv_wrapper_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_conv_wrapper_s8_get_buffer_size_mve): Convolution — Get size of additional buffer required by armdepthwiseconvwrappers8() for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_depthwise_convolve_weight_sum](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_depthwise_convolve_weight_sum): Convolution — Pre-computes per-channel weight sums for a depthwise convolution. — group: Convolution — headers: arm_nnfunctions.h
- [arm_depthwise_nhwc_conv_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f16): Convolution — Depthwise convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_nhwc_conv_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f16_acc16): Convolution — Depthwise convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_depthwise_nhwc_conv_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_depthwise_nhwc_conv_f32): Convolution — Depthwise convolution, NHWC layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_dequantize_f16_f32](/ns-cmsis-nn/reference/api/heliacore/quantization/#arm_dequantize_f16_f32): Pooling, softmax, quantization — Widen a float16 vector to float32. — group: Pooling, softmax, quantization — dtypes: f16, f32 — headers: arm_nnfunctions_flt.h
- [arm_dequantize_s16_f32](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dequantize_s16_f32): Pooling, softmax, quantization — Dequantize an int16t array back to floating-point format. — group: Pooling, softmax, quantization — dtypes: s16, f32 — headers: arm_nnfunctions.h
- [arm_dequantize_s8_f32](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dequantize_s8_f32): Pooling, softmax, quantization — Dequantize an int8t array back to floating-point format. — group: Pooling, softmax, quantization — dtypes: s8, f32 — headers: arm_nnfunctions.h
- [arm_dynamic_update_slice_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dynamic_update_slice_s16): Data movement — Update a slice of an int16 operand tensor at runtime-determined indices. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_dynamic_update_slice_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_dynamic_update_slice_s8): Data movement — Update a slice of an int8 operand tensor at runtime-determined indices. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_elementwise_add_broadcast_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_broadcast_f16): Elementwise — Elementwise add with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_add_broadcast_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_broadcast_f32): Elementwise — Elementwise add with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_add_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_f16): Elementwise — Elementwise add with optional output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_add_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_f32): Elementwise — Elementwise add with optional output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_add_fp16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_add_fp16): Elementwise — Legacy float16 elementwise add with fused clamp, kept only for source compatibility with callers that predate armelementwiseaddf16(). — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_add_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_add_s16): Elementwise — s16 elementwise add of two vectors — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_elementwise_add_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_add_s8): Elementwise — s8 elementwise add of two vectors — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_elementwise_mul_broadcast_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_broadcast_f16): Elementwise — Elementwise multiply with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_mul_broadcast_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_broadcast_f32): Elementwise — Elementwise multiply with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_mul_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_f16): Elementwise — Elementwise multiply with optional output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_mul_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_mul_f32): Elementwise — Elementwise multiply with optional output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_mul_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_mul_s16): Elementwise — s16 elementwise multiplication — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_elementwise_mul_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_mul_s8): Elementwise — s8 elementwise multiplication — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_elementwise_prelu_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_prelu_s16): Elementwise — Elementwise S16 PReLU activation function. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_elementwise_prelu_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_prelu_s8): Elementwise — Elementwise S8 PReLU activation function. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_elementwise_squared_difference_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_squared_difference_f16): Elementwise — Elementwise squared difference of two float16 vectors. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_squared_difference_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_squared_difference_s16): Elementwise — s16 elementwise squared difference of two vectors. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_elementwise_squared_difference_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_squared_difference_s8): Elementwise — s8 elementwise squared difference of two vectors. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_elementwise_sub_broadcast_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_broadcast_f16): Elementwise — Elementwise subtract with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_sub_broadcast_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_broadcast_f32): Elementwise — Elementwise subtract with TensorFlow Lite NHWC broadcasting and an output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_sub_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_f16): Elementwise — Elementwise subtract with optional output clamp. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_sub_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_elementwise_sub_f32): Elementwise — Elementwise subtract with optional output clamp. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_elementwise_sub_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_sub_s16): Elementwise — s16 elementwise subtract of two vectors — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_elementwise_sub_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_elementwise_sub_s8): Elementwise — s8 elementwise subtract of two vectors — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_equal_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_equal_s16): Reduction and comparison — s16 elementwise equality comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_equal_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_equal_s8): Reduction and comparison — s8 elementwise equality comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_f16](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16): Fully connected — Fully connected layer, dispatch by layout. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16_acc16): Fully connected — Fully connected layer, dispatch by layout. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f16_get_buffer_size): Fully connected — Get the temporary buffer size required by the fully connected layer. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_f32](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f32): Fully connected — Fully connected layer, dispatch by layout. — group: Fully connected — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_f32_get_buffer_size): Fully connected — Get the temporary buffer size required by the fully connected layer. — group: Fully connected — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_nhwc_f16](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f16): Fully connected — Fully connected layer, NHWC layout. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_nhwc_f16_acc16](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f16_acc16): Fully connected — Fully connected layer, NHWC layout. — group: Fully connected — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_nhwc_f32](/ns-cmsis-nn/reference/api/heliacore/fc/#arm_fully_connected_nhwc_f32): Fully connected — Fully connected layer, NHWC layout. — group: Fully connected — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_fully_connected_per_channel_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16): Fully connected — Basic s16 Fully Connected function using per channel quantization. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_per_channel_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size): Fully connected — Get size of additional buffer required by armfullyconnectedperchannels16(). — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_per_channel_s16_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size_dsp): Fully connected — Get size of additional buffer required by armfullyconnectedperchannels16() for processors with DSP extension. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_per_channel_s16_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s16_get_buffer_size_mve): Fully connected — Get size of additional buffer required by armfullyconnectedperchannels16() for Arm(R) Helium Architecture case. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_per_channel_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_per_channel_s8): Fully connected — Basic s8 Fully Connected function using per channel quantization. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16): Fully connected — Basic s16 Fully Connected function. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_s16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size): Fully connected — Get size of additional buffer required by armfullyconnecteds16(). — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_s16_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size_dsp): Fully connected — Get size of additional buffer required by armfullyconnecteds16() for processors with DSP extension. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_s16_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s16_get_buffer_size_mve): Fully connected — Get size of additional buffer required by armfullyconnecteds16() for Arm(R) Helium Architecture case. — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_s4](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s4): Fully connected — Basic s4 Fully Connected function. — group: Fully connected — dtypes: s4 — headers: arm_nnfunctions.h
- [arm_fully_connected_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8): Fully connected — Basic s8 Fully Connected function. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size): Fully connected — Get size of additional buffer required by armfullyconnecteds8(). — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size_dsp): Fully connected — Get size of additional buffer required by armfullyconnecteds8() for processors with DSP extension. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_s8_get_buffer_size_mve): Fully connected — Get size of additional buffer required by armfullyconnecteds8() for Arm(R) Helium Architecture case. — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_fully_connected_wrapper_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_wrapper_s16): Fully connected — s16 Fully Connected layer wrapper function — group: Fully connected — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_fully_connected_wrapper_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_fully_connected_wrapper_s8): Fully connected — s8 Fully Connected layer wrapper function — group: Fully connected — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_gather_f16](/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_f16): Data movement — Gather contiguous slices along an axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_gather_f32](/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_f32): Data movement — Gather contiguous slices along an axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_gather_nd_f16](/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_nd_f16): Data movement — Gather contiguous slices using coordinate tuples. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_gather_nd_f32](/ns-cmsis-nn/reference/api/heliacore/gather/#arm_gather_nd_f32): Data movement — Gather contiguous slices using coordinate tuples. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_gather_nd_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_nd_s16): Data movement — Gathernd slices for int16 tensors. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_gather_nd_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_nd_s8): Data movement — Gathernd slices for int8 tensors. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_gather_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_s16): Data movement — Gather elements along an axis for int16 tensors. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_gather_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_gather_s8): Data movement — Gather elements along an axis for int8 tensors. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_greater_equal_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_equal_s16): Reduction and comparison — s16 elementwise greater-or-equal comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_greater_equal_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_equal_s8): Reduction and comparison — s8 elementwise greater-or-equal comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_greater_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_s16): Reduction and comparison — s16 elementwise greater-than comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_greater_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_greater_s8): Reduction and comparison — s8 elementwise greater-than comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_gru_unidirectional_f16](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f16): Sequence — Unidirectional GRU layer for float16 input, output and state. — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_gru_unidirectional_f16_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f16_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armgruunidirectionalf16(). — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_gru_unidirectional_f32](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f32): Sequence — Unidirectional GRU layer for float32 input, output and state. — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_gru_unidirectional_f32_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_gru_unidirectional_f32_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armgruunidirectionalf32(). — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_hard_swish_compat_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_compat_s8): Activation — S8 Hard-Swish activation function (compatibility version). — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_hard_swish_f16](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_hard_swish_f16): Activation — Hard swish activation for float16 data. — group: Activation — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_hard_swish_f32](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_hard_swish_f32): Activation — Hard swish activation for float32 data. — group: Activation — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_hard_swish_precise_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_precise_s16): Activation — S16 Hard-Swish activation function (precise version). — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_hard_swish_precise_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_hard_swish_precise_s8): Activation — S8 Hard-Swish activation function (precise version). — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_leaky_relu_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_leaky_relu_s16): Activation — S16 Leaky ReLU activation function. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_leaky_relu_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_leaky_relu_s8): Activation — S8 Leaky ReLU activation function. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_less_equal_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_equal_s16): Reduction and comparison — s16 elementwise less-or-equal comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_less_equal_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_equal_s8): Reduction and comparison — s8 elementwise less-or-equal comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_less_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_s16): Reduction and comparison — s16 elementwise less-than comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_less_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_less_s8): Reduction and comparison — s8 elementwise less-than comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_logistic_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_logistic_s16): Activation — Logistic activation function for s16. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_f16](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16): Sequence — Unidirectional LSTM inference. — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_f16_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armlstmunidirectionalf16(). — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_f16_temp2_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f16_temp2_get_buffer_size): Sequence — Get size of the temp2 scratch buffer required by armlstmunidirectionalf16(). — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_f32](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32): Sequence — Unidirectional LSTM inference. — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_f32_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armlstmunidirectionalf32(). — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_f32_temp2_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/lstm/#arm_lstm_unidirectional_f32_temp2_get_buffer_size): Sequence — Get size of the temp2 scratch buffer required by armlstmunidirectionalf32(). — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_lstm_unidirectional_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16): Sequence — LSTM unidirectional function with 16 bit input and output and 16 bit gate output, 64 bit bias. — group: Sequence — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_s16_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armlstmunidirectionals16(). — group: Sequence — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_s16_temp2_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s16_temp2_get_buffer_size): Sequence — Get size of the temp2 scratch buffer required by armlstmunidirectionals16(). — group: Sequence — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8): Sequence — LSTM unidirectional function with 8 bit input and output and 16 bit gate output, 32 bit bias. — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_s8_temp1_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8_temp1_get_buffer_size): Sequence — Get size of the temp1 scratch buffer required by armlstmunidirectionals8(). — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_lstm_unidirectional_s8_temp2_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_lstm_unidirectional_s8_temp2_get_buffer_size): Sequence — Get size of the temp2 scratch buffer required by armlstmunidirectionals8(). — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_max_pool_f16](/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_max_pool_f16): Pooling, softmax, quantization — Max pooling. — group: Pooling, softmax, quantization — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_max_pool_f32](/ns-cmsis-nn/reference/api/heliacore/pooling/#arm_max_pool_f32): Pooling, softmax, quantization — Max pooling. — group: Pooling, softmax, quantization — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_max_pool_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_max_pool_s16): Pooling, softmax, quantization — s16 max pooling function. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_max_pool_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_max_pool_s8): Pooling, softmax, quantization — s8 max pooling function. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_maximum_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_maximum_f16): Elementwise — 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. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_maximum_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_maximum_f32): Elementwise — 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. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_maximum_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_maximum_s16): Elementwise — s16 elementwise maximum w/ support for broadcasting and scalar inputs. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_maximum_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_maximum_s8): Elementwise — s8 elementwise maximum w/ support for broadcasting and scalar inputs. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_mean_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mean_s16): Reduction and comparison — Computes the mean of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_mean_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mean_s8): Reduction and comparison — Computes the mean of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_minimum_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_minimum_f16): Elementwise — 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. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_minimum_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_minimum_f32): Elementwise — 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. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_minimum_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_minimum_s16): Elementwise — s16 elementwise minimum w/ support for broadcasting and scalar inputs. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_minimum_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_minimum_s8): Elementwise — s8 elementwise minimum w/ support for broadcasting and scalar inputs. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_mirror_pad_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mirror_pad_s16): Data movement — Mirror-pad an int16 tensor (REFLECT or SYMMETRIC mode). — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_mirror_pad_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mirror_pad_s8): Data movement — Mirror-pad an int8 tensor (REFLECT or SYMMETRIC mode). — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_mul_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_s16): Elementwise — s16 elementwise multiplication of two tensors with support for broadcasting. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_mul_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_s8): Elementwise — s8 elementwise multiplication of two tensors with support for broadcasting. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_mul_scalar_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_scalar_s16): Elementwise — s16 elementwise multiplication of scalar and vector — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_mul_scalar_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_mul_scalar_s8): Elementwise — s8 elementwise multiplication of scalar and vector — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_nn_abs_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_abs_f16): Elementwise — Elementwise absolute value. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_nn_abs_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_abs_f32): Elementwise — Elementwise absolute value. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_nn_activation_f16](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_nn_activation_f16): Activation — Elementwise activation. — group: Activation — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_nn_activation_f32](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_nn_activation_f32): Activation — Elementwise activation. — group: Activation — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_nn_activation_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_nn_activation_s16): Activation — s16 neural network activation function using direct table look-up — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_nn_fill_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_fill_f16): Data movement — Fill a float16 vector with one value; bit copy of value, NaN payload included. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_nn_fill_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_fill_f32): Data movement — Fill a float32 vector with one value. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_nn_mean_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_nn_mean_f16): Reduction and comparison — Computes the mean of a float16 tensor along the specified axes. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_nn_mean_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_nn_mean_f32): Reduction and comparison — Computes the mean of a float32 tensor along the specified axes. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_nn_sqrt_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_sqrt_f16): Elementwise — Elementwise square root of a float16 tensor. — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_nn_sqrt_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_nn_sqrt_f32): Elementwise — Elementwise square root. — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_not_equal_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_not_equal_s16): Reduction and comparison — s16 elementwise inequality comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_not_equal_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_not_equal_s8): Reduction and comparison — s8 elementwise inequality comparison with support for broadcasting. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_pack_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_pack_f16): Data movement — Stack float32 tensors of equal shape along a new axis (TFLite PACK). — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_pack_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_pack_f32): Data movement — Stack float32 tensors of equal shape along a new axis (TFLite PACK). — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_pad_f16](/ns-cmsis-nn/reference/api/heliacore/pad/#arm_pad_f16): Data movement — Pad a tensor with a constant value. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_pad_f32](/ns-cmsis-nn/reference/api/heliacore/pad/#arm_pad_f32): Data movement — Pad a tensor with a constant value. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_pad_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_pad_s16): Data movement — Expands the size of the input by adding constant values before and after the data, in all dimensions. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_pad_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_pad_s8): Data movement — Expands the size of the input by adding constant values before and after the data, in all dimensions. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_prelu_f16](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_prelu_f16): Activation — Parametric ReLU for float32 data. — group: Activation — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_prelu_f32](/ns-cmsis-nn/reference/api/heliacore/acti/#arm_prelu_f32): Activation — Parametric ReLU for float32 data. — group: Activation — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_prelu_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_s16): Activation — S16 PReLU activation function. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_prelu_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_s8): Activation — S8 PReLU activation function. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_prelu_scalar_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_scalar_s16): Activation — Scalar S16 PReLU activation function. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_prelu_scalar_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_prelu_scalar_s8): Activation — Scalar S8 PReLU activation function. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_quantize_f32_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_quantize_f32_s16): Pooling, softmax, quantization — Quantize a floating-point array into int16t format. — group: Pooling, softmax, quantization — dtypes: f32, s16 — headers: arm_nnfunctions.h
- [arm_quantize_f32_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_quantize_f32_s8): Pooling, softmax, quantization — Quantize a floating-point array into int8t format. — group: Pooling, softmax, quantization — dtypes: f32, s8 — headers: arm_nnfunctions.h
- [arm_reduce_max_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_max_f16): Reduction and comparison — Reduces a f16 NHWC tensor to its maximum along a binary axis mask. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_reduce_max_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_max_f32): Reduction and comparison — Reduces a f32 NHWC tensor to its maximum along a binary axis mask. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_reduce_max_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_max_s16): Reduction and comparison — Computes the max of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_reduce_max_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_max_s8): Reduction and comparison — Computes the max of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_reduce_min_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_min_f16): Reduction and comparison — Reduces a f16 NHWC tensor to its minimum along a binary axis mask. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_reduce_min_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_min_f32): Reduction and comparison — Reduces a f32 NHWC tensor to its minimum along a binary axis mask. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_reduce_min_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_min_s16): Reduction and comparison — Computes the min of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_reduce_min_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reduce_min_s8): Reduction and comparison — Computes the min of the input tensor along the specified axis. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_reduce_sum_f16](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_sum_f16): Reduction and comparison — Computes the sum of the input tensor along the specified axes. — group: Reduction and comparison — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_reduce_sum_f32](/ns-cmsis-nn/reference/api/heliacore/reduction/#arm_reduce_sum_f32): Reduction and comparison — Computes the sum of the input tensor along the specified axes. — group: Reduction and comparison — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_relu_generic_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_generic_s16): Activation — S16 ReLU activation function (generic version) This generic version allows to set custom lower and upper bounds to implement ReLU6, etc. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_relu_generic_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_generic_s8): Activation — S8 ReLU activation function (generic version) This generic version allows to set custom lower and upper bounds to implement ReLU6, etc. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_relu_q15](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_q15): Activation — Q15 RELU function. — group: Activation — dtypes: q15 — headers: arm_nnfunctions.h
- [arm_relu_q7](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_q7): Activation — Q7 RELU function. — group: Activation — dtypes: q7 — headers: arm_nnfunctions.h
- [arm_relu_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_s16): Activation — S16 ReLU activation function lower and upper bounds are quantized representations of 0 and 32767. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_relu_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu_s8): Activation — S8 ReLU activation function lower and upper bounds are quantized representations of 0 and 127. — group: Activation — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_relu6_q7](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_relu6_q7): Activation — Q7 RELU6 function. — group: Activation — dtypes: q7 — headers: arm_nnfunctions.h
- [arm_requantize_s16_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_requantize_s16_s16): Pooling, softmax, quantization — Requantize an int16t array to another int16t range with a different scale. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_requantize_s8_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_requantize_s8_s8): Pooling, softmax, quantization — Requantize an int8t array to another int8t range with a different scale. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_reshape_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_reshape_f16): Data movement — Reshape by copying data without changing element order. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_reshape_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_reshape_f32): Data movement — Reshape by copying data without changing element order. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_reshape_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reshape_s8): Data movement — Reshape a s8 vector into another with different shape. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_resize_nearest_neighbor_f16](/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f16): Data movement — Nearest-neighbor resize of a float32 NHWC tensor. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_resize_nearest_neighbor_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f16_get_buffer_size): Data movement — Scratch size in bytes for armresizenearestneighborf32() / armresizenearestneighborf16(). — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_resize_nearest_neighbor_f32](/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f32): Data movement — Nearest-neighbor resize of a float32 NHWC tensor. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_resize_nearest_neighbor_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/reshape/#arm_resize_nearest_neighbor_f32_get_buffer_size): Data movement — Scratch size in bytes for armresizenearestneighborf32() / armresizenearestneighborf16(). — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_resize_nearest_neighbor_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_resize_nearest_neighbor_s16): Data movement — Nearest neighbor resize function for s16 data. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_resize_nearest_neighbor_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_resize_nearest_neighbor_s8): Data movement — Nearest neighbor resize function for s8 data. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_reverse_sequence_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reverse_sequence_s16): Data movement — Reverse variable-length sequences along a dimension for int16. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_reverse_sequence_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_reverse_sequence_s8): Data movement — Reverse variable-length sequences along a dimension for int8. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_rsqrt_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_rsqrt_f16): Elementwise — Elementwise reciprocal square root of a float16 tensor, 1 / sqrt(x). — group: Elementwise — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_rsqrt_f32](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_rsqrt_f32): Elementwise — Elementwise reciprocal square root, 1 / sqrt(x). — group: Elementwise — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_rsqrt_s16_per_op](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_rsqrt_s16_per_op): Elementwise — INT16 reciprocal square root using a per-operator LUT. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_rsqrt_s16_universal](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_rsqrt_s16_universal): Elementwise — INT16 reciprocal square root using a shared universal LUT. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_scatter_nd_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_scatter_nd_s16): Data movement — Scatter updates into a zero-initialized output tensor for int16. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_scatter_nd_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_scatter_nd_s8): Data movement — Scatter updates into a zero-initialized output tensor for int8. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_select_v2_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_select_v2_s16): Elementwise — SELECTV2 with broadcast for int16 tensors. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_select_v2_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_select_v2_s8): Elementwise — SELECTV2 with broadcast for int8 tensors. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_softmax_f16](/ns-cmsis-nn/reference/api/heliacore/softmax/#arm_softmax_f16): Pooling, softmax, quantization — Softmax using the float-native API signature. — group: Pooling, softmax, quantization — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_softmax_f32](/ns-cmsis-nn/reference/api/heliacore/softmax/#arm_softmax_f32): Pooling, softmax, quantization — Softmax using the float-native API signature. — group: Pooling, softmax, quantization — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_softmax_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s16): Pooling, softmax, quantization — S16 softmax function. — group: Pooling, softmax, quantization — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_softmax_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s8): Pooling, softmax, quantization — S8 softmax function. — group: Pooling, softmax, quantization — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_softmax_s8_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_s8_s16): Pooling, softmax, quantization — S8 to s16 softmax function. — group: Pooling, softmax, quantization — dtypes: s8, s16 — headers: arm_nnfunctions.h
- [arm_softmax_u8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_softmax_u8): Pooling, softmax, quantization — U8 softmax function. — group: Pooling, softmax, quantization — dtypes: u8 — headers: arm_nnfunctions.h
- [arm_space_to_batch_nd_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_batch_nd_s16): Data movement — Space to Batch ND function for s16 data type. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_space_to_batch_nd_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_batch_nd_s8): Data movement — Space to Batch ND function for s8 data type. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_space_to_depth_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_depth_s16): Data movement — Space to Depth function for s16 data type. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_space_to_depth_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_space_to_depth_s8): Data movement — Space to Depth function for s8 data type. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_split_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_split_f16): Data movement — Split a float32 tensor of any rank into several tensors along one axis. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_split_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_split_f32): Data movement — Split a float32 tensor of any rank into several tensors along one axis. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_split_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_split_s16): Data movement — int16/uint16 split function to be used for splitting a tensor into multiple tensors along the target axis — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_split_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_split_s8): Data movement — int8/uint8 split function to be used for splitting a tensor into multiple tensors along the target axis — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_sqrt_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s16): Elementwise — s16 elementwise square root using piecewise LUT with linear interpolation — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_sqrt_s16_tablefree](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s16_tablefree): Elementwise — s16 elementwise square root without a lookup table — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_sqrt_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sqrt_s8): Elementwise — s8 elementwise square root — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_squared_difference_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_s16): Elementwise — s16 elementwise squared difference of two tensors with support for broadcasting. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_squared_difference_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_s8): Elementwise — s8 elementwise squared difference of two tensors with support for broadcasting. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_squared_difference_scalar_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_scalar_s16): Elementwise — s16 elementwise squared difference of scalar and vector. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_squared_difference_scalar_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_squared_difference_scalar_s8): Elementwise — s8 elementwise squared difference of scalar and vector. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_strided_slice_f16](/ns-cmsis-nn/reference/api/heliacore/groupelementwise/#arm_strided_slice_f16): Data movement — Strided slice for float32 data (pure copy, TensorFlow Lite compatible). — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_strided_slice_f32](/ns-cmsis-nn/reference/api/heliacore/stridedslice/#arm_strided_slice_f32): Data movement — Strided slice for float32 data (pure copy, TensorFlow Lite compatible). — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_strided_slice_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s16): Data movement — Strided slice function for int16 data. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_strided_slice_s32](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s32): Data movement — Strided slice function for int32 data. — group: Data movement — dtypes: s32 — headers: arm_nnfunctions.h
- [arm_strided_slice_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_strided_slice_s8): Data movement — Strided slice function for int8 data. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_sub_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_s16): Elementwise — s16 elementwise subtraction of two tensors with support for broadcasting. — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_sub_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_s8): Elementwise — s8 elementwise subtraction of two tensors with support for broadcasting. — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_sub_scalar_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_scalar_s16): Elementwise — s16 elementwise subtract of scalar and vector (scalar - vector) — group: Elementwise — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_sub_scalar_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_sub_scalar_s8): Elementwise — s8 elementwise subtract of scalar and vector (scalar - vector) — group: Elementwise — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_f16](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16): Sequence — Stateful singular value decomposition filter, float16 variant. — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_svdf_f16_input_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16_input_ctx_get_buffer_size): Sequence — Get size of the inputctx staging buffer required by armsvdff16(). — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_svdf_f16_output_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f16_output_ctx_get_buffer_size): Sequence — Get size of the outputctx staging buffer required by armsvdff16(). — group: Sequence — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_svdf_f32](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32): Sequence — Stateful singular value decomposition filter. — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_svdf_f32_input_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32_input_ctx_get_buffer_size): Sequence — Get size of the inputctx staging buffer required by armsvdff32(). — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_svdf_f32_output_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/svdf/#arm_svdf_f32_output_ctx_get_buffer_size): Sequence — Get size of the outputctx staging buffer required by armsvdff32(). — group: Sequence — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_svdf_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8): Sequence — s8 SVDF function with 8 bit state tensor and 8 bit time weights — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size): Sequence — Get size of the kernel-sum buffer required by armsvdfs8(). — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_s8_get_buffer_size_dsp](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size_dsp): Sequence — Get size of the kernel-sum buffer required by armsvdfs8() for processors with DSP extension. — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_get_buffer_size_mve): Sequence — Get size of the kernel-sum buffer required by armsvdfs8() for Arm(R) Helium Architecture case. — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_s8_input_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_input_ctx_get_buffer_size): Sequence — Get size of the inputctx staging buffer required by armsvdfs8(). — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_s8_output_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_s8_output_ctx_get_buffer_size): Sequence — Get size of the outputctx staging buffer required by armsvdfs8(). — group: Sequence — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_svdf_state_s16_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8): Sequence — s8 SVDF function with 16 bit state tensor and 16 bit time weights — group: Sequence — dtypes: s16, s8 — headers: arm_nnfunctions.h
- [arm_svdf_state_s16_s8_input_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8_input_ctx_get_buffer_size): Sequence — Get size of the inputctx staging buffer required by armsvdfstates16s8(). — group: Sequence — dtypes: s16, s8 — headers: arm_nnfunctions.h
- [arm_svdf_state_s16_s8_output_ctx_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_svdf_state_s16_s8_output_ctx_get_buffer_size): Sequence — Get size of the outputctx staging buffer required by armsvdfstates16s8(). — group: Sequence — dtypes: s16, s8 — headers: arm_nnfunctions.h
- [arm_tanh_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tanh_s16): Activation — Tanh activation function for s16. — group: Activation — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_tile_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tile_s16): Data movement — Tile an int16 tensor along each dimension. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_tile_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_tile_s8): Data movement — Tile an int8 tensor along each dimension. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_conv_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16): Convolution — Transpose convolution, dispatch by layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_f16_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16_get_buffer_size): Convolution — Get the temporary buffer size required by transpose convolution. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_f16_get_reverse_conv_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f16_get_reverse_conv_buffer_size): Convolution — Get the reverse-convolution workspace size used by transpose convolution helpers. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32): Convolution — Transpose convolution, dispatch by layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_f32_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32_get_buffer_size): Convolution — Get the temporary buffer size required by transpose convolution. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_f32_get_reverse_conv_buffer_size](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_f32_get_reverse_conv_buffer_size): Convolution — Get the reverse-convolution workspace size used by transpose convolution helpers. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_nhwc_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_nhwc_f16): Convolution — Transpose convolution, NHWC layout. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_nhwc_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_nhwc_f32): Convolution — Transpose convolution, NHWC layout. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8): Convolution — Basic s8 transpose convolution function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_conv_s8_get_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_buffer_size): Convolution — Get the required buffer size for ctx in s8 transpose conv function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_conv_s8_get_buffer_size_mve](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_buffer_size_mve): Convolution — Get size of additional buffer required by armtransposeconvs8() for Arm(R) Helium Architecture case. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_conv_s8_get_reverse_conv_buffer_size](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_s8_get_reverse_conv_buffer_size): Convolution — Get the required buffer size for outputctx in s8 transpose conv function. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_conv_wrapper_f16](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_wrapper_f16): Convolution — Transpose convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_wrapper_f32](/ns-cmsis-nn/reference/api/heliacore/nnconv/#arm_transpose_conv_wrapper_f32): Convolution — Transpose convolution wrapper using the CMSIS-NN baseline path. — group: Convolution — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_conv_wrapper_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_conv_wrapper_s8): Convolution — Wrapper to select optimal transposed convolution algorithm depending on parameters. — group: Convolution — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_transpose_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_transpose_f16): Data movement — Transpose a floating-point tensor. — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_transpose_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_transpose_f32): Data movement — Transpose a floating-point tensor. — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_transpose_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_s16): Data movement — Basic s16 transpose function. — group: Data movement — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_transpose_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_transpose_s8): Data movement — Basic transpose function. — group: Data movement — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_unpack_f16](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_unpack_f16): Data movement — Unstack a float32 tensor along one axis into inputshape\[axis\] tensors (TFLite UNPACK). — group: Data movement — dtypes: f16 — headers: arm_nnfunctions_flt.h
- [arm_unpack_f32](/ns-cmsis-nn/reference/api/heliacore/nnsupport/#arm_unpack_f32): Data movement — Unstack a float32 tensor along one axis into inputshape\[axis\] tensors (TFLite UNPACK). — group: Data movement — dtypes: f32 — headers: arm_nnfunctions_flt.h
- [arm_vector_sum_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_vector_sum_s8): Reduction and comparison — Calculate the sum of each row in vectordata, multiply by lhsoffset and optionally add s32 biasdata. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h
- [arm_vector_sum_s8_s64](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_vector_sum_s8_s64): Reduction and comparison — Calculate the sum of each row in vectordata, multiply by lhsoffset and optionally add s64 biasdata. — group: Reduction and comparison — dtypes: s8, s64 — headers: arm_nnfunctions.h
- [arm_where_s16](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_where_s16): Reduction and comparison — WHERE operator: return coordinates of non-zero elements in condition (int16). — group: Reduction and comparison — dtypes: s16 — headers: arm_nnfunctions.h
- [arm_where_s8](/ns-cmsis-nn/reference/api/heliacore/arm_nnfunctions/#arm_where_s8): Reduction and comparison — WHERE operator: return coordinates of non-zero elements in condition. — group: Reduction and comparison — dtypes: s8 — headers: arm_nnfunctions.h