Edge AI Add-on API 2.2.0
Loading...
Searching...
No Matches
nrf_edgeai_runtime.h File Reference

Go to the source code of this file.

Functions

nrf_edgeai_err_t nrf_edgeai_init (nrf_edgeai_t *p_edgeai)
 Set up the internal components of the Edge AI runtime.
 
nrf_edgeai_err_t nrf_edgeai_feed_inputs (nrf_edgeai_t *p_edgeai, void *p_input_values, uint16_t num_values)
 Feed raw input data to prepare it for signal processing & model inference.
 
nrf_edgeai_err_t nrf_edgeai_run_inference (nrf_edgeai_t *p_edgeai)
 Running live input features into a machine learning algorithm (or “ML/NN model”) to inference an output.
 
nrf_edgeai_input_type_t nrf_edgeai_input_type (const nrf_edgeai_t *p_edgeai)
 Get neural network input data type nrf_edgeai_input_type_t.
 
uint16_t nrf_edgeai_uniq_inputs_num (const nrf_edgeai_t *p_edgeai)
 Get number of unique input features on which the model was trained, e.g for features {x, y, z} -> number of unique input features = 3.
 
uint16_t nrf_edgeai_input_window_size (const nrf_edgeai_t *p_edgeai)
 Get input features window size in feature samples(vectors), e.g for input window {x0, y0, z0, ..., xn, yn, zn} -> window size = n.
 
uint8_t nrf_edgeai_input_subwindows_num (const nrf_edgeai_t *p_edgeai)
 Get number of subwindows in the input window.
 
uint16_t nrf_edgeai_model_outputs_num (const nrf_edgeai_t *p_edgeai)
 Get number of model outputs (predicted targets)
 
nrf_edgeai_model_type_t nrf_edgeai_model_type (const nrf_edgeai_t *p_edgeai)
 Get model type nrf_edgeai_model_type_t.
 
nrf_edgeai_model_task_t nrf_edgeai_model_task (const nrf_edgeai_t *p_edgeai)
 Get model task nrf_edgeai_model_task_t.
 
const char * nrf_edgeai_solution_id_str (const nrf_edgeai_t *p_edgeai)
 Get solution ID in string format.
 
nrf_edgeai_rt_version_t nrf_edgeai_solution_runtime_version (const nrf_edgeai_t *p_edgeai)
 Get solution runtime version.
 
nrf_edgeai_rt_version_t nrf_edgeai_runtime_version (void)
 Get Edge AI runtime library version.
 
bool nrf_edgeai_is_runtime_compatible (const nrf_edgeai_t *p_edgeai)
 Check if the Edge AI runtime library version is compatible with the solution runtime version.