Axon Neural Network processing library

Typedefs
typedef struct nrf_axon_nn_compiled_model_tag_s	nrf_nn_axon_model_t

Functions
int8_t	nrf_nn_axon_init (const nrf_nn_axon_model_t *p_nn)
	Initialize the Axon NN inference engine.
int8_t	nrf_nn_axon_deinit (const nrf_nn_axon_model_t *p_nn)
	Deinitialize the Axon NN inference engine and release any resources.
int8_t	nrf_nn_axon_init_persistent_vars (const nrf_nn_axon_model_t *p_nn)
	Initialize all the persistent var buffers in a streaming-style model (with VarHandle/ReadVariable/AssignVariable). Should be called at the start of each streaming session. Harmless to call for non-streaming style models.
void	nrf_nn_axon_scale_inputs (const nrf_nn_axon_model_t *p_nn, const flt32_t *p_input_features, size32_t input_features_num, int8_t *p_scaled_features)
	Scale floating point input features to quantized int8 input tensor format.
int8_t	nrf_nn_axon_run_inference (const nrf_nn_axon_model_t *p_nn, const int8_t *p_input_features, int8_t *p_output_buffer)
	Run inference on an Axon NN model.
void	nrf_nn_axon_dequantize_outputs (const nrf_nn_axon_model_t *p_nn, const int8_t *p_output_buffer, uint16_t outputs_num, flt32_t *p_dequantized_outputs)
	Dequantize the outputs from an Axon NN model.

Detailed Description