API¶

heliaCORE keeps the CMSIS-NN C API shape where compatibility matters, then adds Ambiq-tuned kernels and integration paths for HELIA workloads on Apollo-class targets. Start with the operator families below, filter to the kernels you care about, then drill into exact generated signatures when you need implementation detail.

positioningWhat it isA CMSIS-NN-compatible kernel layer packaged for Ambiq firmware and HELIA runtime integration. compatibilityHow it differsInherited CMSIS-NN APIs stay recognizable while Ambiq coverage fills model-graph gaps around real deployments. integrationHow it fits HELIAUse heliaCORE through neuralSPOT-X or direct CMake/CMSIS-Pack flows depending on your firmware stack.

Browse By Kernel Family¶

computeConvolutionConv2D, depthwise, transpose convolution, wrappers, and buffer helpers. computeFully connectedDense layers, batch matmul paths, and scratch sizing helpers. graph glueElementwiseAdd, sub, mul, square difference, min/max, batch norm, select, and arithmetic glue. selectionReduction and comparisonArgmin/argmax, min/max reductions, comparisons, means, vector sums, and where. activationActivationReLU, LeakyReLU, PReLU, Hard-Swish, Logistic, Tanh, and clamp. layoutData movementPad, reshape, transpose, concatenate, gather, resize, strided slice, tile, broadcast, scatter, mirror pad, and sequence/slice-update utilities. tailPooling, softmax, quantizationClassifier tail APIs plus dtype conversion and requantization utilities. sequenceSequenceLSTM, GRU, and SVDF functions for temporal workloads. indexTypes and supportStructs, enums, files, and low-level support helpers in the full generated index.

For file-by-file browsing, structs, enums, and the complete generated tree, open the generated C API reference.

Convolution Functions¶

112 functions

arm_convolve_1_x_n_f16 arm_convolve_1_x_n_f16_get_buffer_size arm_convolve_1_x_n_f32 arm_convolve_1_x_n_f32_get_buffer_size arm_convolve_1_x_n_nhwc_f16 arm_convolve_1_x_n_nhwc_f32 arm_convolve_1_x_n_s4 arm_convolve_1_x_n_s4_get_buffer_size arm_convolve_1_x_n_s8 arm_convolve_1_x_n_s8_get_buffer_size arm_convolve_1x1_f16 arm_convolve_1x1_f16_get_buffer_size arm_convolve_1x1_f32 arm_convolve_1x1_f32_get_buffer_size arm_convolve_1x1_nhwc_f16 arm_convolve_1x1_nhwc_f32 arm_convolve_1x1_out_s8 arm_convolve_1x1_out_s8_get_buffer_size arm_convolve_1x1_s16_ns_np_nd arm_convolve_1x1_s4 arm_convolve_1x1_s4_fast arm_convolve_1x1_s4_fast_get_buffer_size arm_convolve_1x1_s8 arm_convolve_1x1_s8_fast arm_convolve_1x1_s8_fast_get_buffer_size arm_convolve_even_s4 arm_convolve_even_s4_get_buffer_size arm_convolve_f16 arm_convolve_f16_get_buffer_size arm_convolve_f32 arm_convolve_f32_get_buffer_size arm_convolve_nhwc_f16 arm_convolve_nhwc_f32 arm_convolve_s16 arm_convolve_s16_fast_small_kernel arm_convolve_s16_get_buffer_size arm_convolve_s16_group_ch_mult_1 arm_convolve_s4 arm_convolve_s4_get_buffer_size arm_convolve_s8 arm_convolve_s8_get_buffer_size arm_convolve_s8_get_buffer_size_mve arm_convolve_s8_get_weights_sum_size arm_convolve_weight_sum arm_convolve_wrapper_f16 arm_convolve_wrapper_f16_get_buffer_size arm_convolve_wrapper_f32 arm_convolve_wrapper_f32_get_buffer_size arm_convolve_wrapper_s16 arm_convolve_wrapper_s16_get_buffer_size arm_convolve_wrapper_s16_get_buffer_size_dsp arm_convolve_wrapper_s16_get_buffer_size_mve arm_convolve_wrapper_s4 arm_convolve_wrapper_s4_get_buffer_size arm_convolve_wrapper_s4_get_buffer_size_dsp arm_convolve_wrapper_s4_get_buffer_size_mve arm_convolve_wrapper_s8 arm_convolve_wrapper_s8_get_buffer_size arm_convolve_wrapper_s8_get_buffer_size_dsp arm_convolve_wrapper_s8_get_buffer_size_mve arm_depthwise_conv_3x3_s8 arm_depthwise_conv_f16 arm_depthwise_conv_f16_get_buffer_size arm_depthwise_conv_f32 arm_depthwise_conv_f32_get_buffer_size arm_depthwise_conv_fast_s16 arm_depthwise_conv_fast_s16_get_buffer_size arm_depthwise_conv_input_index_nhwc arm_depthwise_conv_output_index_nhwc arm_depthwise_conv_s16 arm_depthwise_conv_s4 arm_depthwise_conv_s4_opt arm_depthwise_conv_s4_opt_get_buffer_size arm_depthwise_conv_s8 arm_depthwise_conv_s8_opt arm_depthwise_conv_s8_opt_get_buffer_size arm_depthwise_conv_s8_opt_get_buffer_size_dsp arm_depthwise_conv_s8_opt_get_buffer_size_mve arm_depthwise_conv_wrapper_f16 arm_depthwise_conv_wrapper_f16_get_buffer_size arm_depthwise_conv_wrapper_f32 arm_depthwise_conv_wrapper_f32_get_buffer_size arm_depthwise_conv_wrapper_s16 arm_depthwise_conv_wrapper_s16_get_buffer_size arm_depthwise_conv_wrapper_s16_get_buffer_size_dsp arm_depthwise_conv_wrapper_s16_get_buffer_size_mve arm_depthwise_conv_wrapper_s4 arm_depthwise_conv_wrapper_s4_get_buffer_size arm_depthwise_conv_wrapper_s4_get_buffer_size_dsp arm_depthwise_conv_wrapper_s4_get_buffer_size_mve arm_depthwise_conv_wrapper_s8 arm_depthwise_conv_wrapper_s8_get_buffer_size arm_depthwise_conv_wrapper_s8_get_buffer_size_dsp arm_depthwise_conv_wrapper_s8_get_buffer_size_mve arm_depthwise_convolve_weight_sum arm_depthwise_nhwc_conv_f16 arm_depthwise_nhwc_conv_f32 arm_transpose_conv_f16 arm_transpose_conv_f16_get_buffer_size arm_transpose_conv_f16_get_reverse_conv_buffer_size arm_transpose_conv_f32 arm_transpose_conv_f32_get_buffer_size arm_transpose_conv_f32_get_reverse_conv_buffer_size arm_transpose_conv_nhwc_f16 arm_transpose_conv_nhwc_f32 arm_transpose_conv_s8 arm_transpose_conv_s8_get_buffer_size arm_transpose_conv_s8_get_buffer_size_mve arm_transpose_conv_s8_get_reverse_conv_buffer_size arm_transpose_conv_wrapper_f16 arm_transpose_conv_wrapper_f32 arm_transpose_conv_wrapper_s8

Fully-Connected Layer Functions¶

31 functions

arm_batch_matmul_f16 arm_batch_matmul_f16_get_buffer_size arm_batch_matmul_f32 arm_batch_matmul_f32_get_buffer_size arm_batch_matmul_s16 arm_batch_matmul_s8 arm_batch_matmul_s8_get_buffer_size arm_batch_matmul_s8_get_buffer_size_dsp arm_batch_matmul_s8_get_buffer_size_mve arm_fully_connected_f16 arm_fully_connected_f16_get_buffer_size arm_fully_connected_f32 arm_fully_connected_f32_get_buffer_size arm_fully_connected_nhwc_f16 arm_fully_connected_nhwc_f32 arm_fully_connected_per_channel_s16 arm_fully_connected_per_channel_s16_get_buffer_size arm_fully_connected_per_channel_s16_get_buffer_size_dsp arm_fully_connected_per_channel_s16_get_buffer_size_mve arm_fully_connected_per_channel_s8 arm_fully_connected_s16 arm_fully_connected_s16_get_buffer_size arm_fully_connected_s16_get_buffer_size_dsp arm_fully_connected_s16_get_buffer_size_mve arm_fully_connected_s4 arm_fully_connected_s8 arm_fully_connected_s8_get_buffer_size arm_fully_connected_s8_get_buffer_size_dsp arm_fully_connected_s8_get_buffer_size_mve arm_fully_connected_wrapper_s16 arm_fully_connected_wrapper_s8

Elementwise Functions¶

69 functions

arm_abs_s16 arm_abs_s8 arm_add_s16 arm_add_s8 arm_add_scalar_s16 arm_add_scalar_s8 arm_batch_norm_f16 arm_batch_norm_f32 arm_elementwise_add_broadcast_f16 arm_elementwise_add_broadcast_f32 arm_elementwise_add_f16 arm_elementwise_add_f32 arm_elementwise_add_fp16 arm_elementwise_add_s16 arm_elementwise_add_s8 arm_elementwise_mul_acc_s16 arm_elementwise_mul_broadcast_f16 arm_elementwise_mul_broadcast_f32 arm_elementwise_mul_f16 arm_elementwise_mul_f32 arm_elementwise_mul_s16 arm_elementwise_mul_s16_batch_offset arm_elementwise_mul_s16_s8 arm_elementwise_mul_s8 arm_elementwise_prelu_s16 arm_elementwise_prelu_s8 arm_elementwise_squared_difference_s16 arm_elementwise_squared_difference_s8 arm_elementwise_sub_broadcast_f16 arm_elementwise_sub_broadcast_f32 arm_elementwise_sub_f16 arm_elementwise_sub_f32 arm_elementwise_sub_s16 arm_elementwise_sub_s8 arm_maximum_f16 arm_maximum_f32 arm_maximum_s16 arm_maximum_s8 arm_minimum_f16 arm_minimum_f32 arm_minimum_s16 arm_minimum_s8 arm_mul_s16 arm_mul_s8 arm_mul_scalar_s16 arm_mul_scalar_s8 arm_nn_abs_f16 arm_nn_abs_f16h arm_nn_abs_f32 arm_nn_sqrt_f16 arm_nn_sqrt_f32 arm_nn_sqrt_special_f16 arm_nn_sqrt_special_f32 arm_rsqrt_f16 arm_rsqrt_f32 arm_rsqrt_s16_per_op arm_rsqrt_s16_universal arm_select_v2_s16 arm_select_v2_s8 arm_sqrt_s16 arm_sqrt_s8 arm_squared_difference_s16 arm_squared_difference_s8 arm_squared_difference_scalar_s16 arm_squared_difference_scalar_s8 arm_sub_s16 arm_sub_s8 arm_sub_scalar_s16 arm_sub_scalar_s8

Data Movement¶

77 functions

arm_batch_to_space_nd_s16 arm_batch_to_space_nd_s8 arm_broadcast_to_s16 arm_broadcast_to_s8 arm_concatenation_f16 arm_concatenation_f16_w arm_concatenation_f16_x arm_concatenation_f16_y arm_concatenation_f16_z arm_concatenation_f32 arm_concatenation_f32_w arm_concatenation_f32_x arm_concatenation_f32_y arm_concatenation_f32_z arm_concatenation_s16 arm_concatenation_s32 arm_concatenation_s8 arm_concatenation_s8_w arm_concatenation_s8_x arm_concatenation_s8_y arm_concatenation_s8_z arm_depth_to_space_s16 arm_depth_to_space_s8 arm_dynamic_update_slice_s16 arm_dynamic_update_slice_s8 arm_gather_f16 arm_gather_f32 arm_gather_nd_f16 arm_gather_nd_f32 arm_gather_nd_s16 arm_gather_nd_s8 arm_gather_s16 arm_gather_s8 arm_mirror_pad_s16 arm_mirror_pad_s8 arm_nn_fill_f16 arm_nn_fill_f32 arm_pack_f16 arm_pack_f32 arm_pad_f16 arm_pad_f32 arm_pad_s16 arm_pad_s8 arm_reshape_f16 arm_reshape_f32 arm_reshape_s8 arm_resize_nearest_neighbor_f16 arm_resize_nearest_neighbor_f16_get_buffer_size arm_resize_nearest_neighbor_f32 arm_resize_nearest_neighbor_f32_get_buffer_size arm_resize_nearest_neighbor_s16 arm_resize_nearest_neighbor_s8 arm_reverse_sequence_s16 arm_reverse_sequence_s8 arm_scatter_nd_s16 arm_scatter_nd_s8 arm_space_to_batch_nd_s16 arm_space_to_batch_nd_s8 arm_space_to_depth_s16 arm_space_to_depth_s8 arm_split_f16 arm_split_f32 arm_split_s16 arm_split_s8 arm_strided_slice_f16 arm_strided_slice_f32 arm_strided_slice_s16 arm_strided_slice_s32 arm_strided_slice_s8 arm_tile_s16 arm_tile_s8 arm_transpose_f16 arm_transpose_f32 arm_transpose_s16 arm_transpose_s8 arm_unpack_f16 arm_unpack_f32

Sequence Functions¶

39 functions

arm_gru_unidirectional_f16 arm_gru_unidirectional_f16_temp1_get_buffer_size arm_gru_unidirectional_f32 arm_gru_unidirectional_f32_temp1_get_buffer_size arm_lstm_unidirectional_f16 arm_lstm_unidirectional_f16_temp1_get_buffer_size arm_lstm_unidirectional_f16_temp2_get_buffer_size arm_lstm_unidirectional_f32 arm_lstm_unidirectional_f32_temp1_get_buffer_size arm_lstm_unidirectional_f32_temp2_get_buffer_size arm_lstm_unidirectional_s16 arm_lstm_unidirectional_s16_temp1_get_buffer_size arm_lstm_unidirectional_s16_temp2_get_buffer_size arm_lstm_unidirectional_s8 arm_lstm_unidirectional_s8_temp1_get_buffer_size arm_lstm_unidirectional_s8_temp2_get_buffer_size arm_nn_gru_step_f16 arm_nn_gru_step_f32 arm_nn_lstm_calculate_gate_s16 arm_nn_lstm_calculate_gate_s8_s16 arm_nn_lstm_step_f16 arm_nn_lstm_step_f32 arm_nn_lstm_step_s16 arm_nn_lstm_step_s8 arm_svdf_f16 arm_svdf_f16_input_ctx_get_buffer_size arm_svdf_f16_output_ctx_get_buffer_size arm_svdf_f32 arm_svdf_f32_input_ctx_get_buffer_size arm_svdf_f32_output_ctx_get_buffer_size arm_svdf_s8 arm_svdf_s8_get_buffer_size arm_svdf_s8_get_buffer_size_dsp arm_svdf_s8_get_buffer_size_mve arm_svdf_s8_input_ctx_get_buffer_size arm_svdf_s8_output_ctx_get_buffer_size arm_svdf_state_s16_s8 arm_svdf_state_s16_s8_input_ctx_get_buffer_size arm_svdf_state_s16_s8_output_ctx_get_buffer_size

Types and Support¶

Structs, enums, files, and lower-level support helpers are intentionally kept in the full generated C API index to avoid duplicating declarations across the grouped operator view and the complete Exhale tree.