oveRTOS C API
Embedded RTOS framework — build system, configuration, and portable C API
Loading...
Searching...
No Matches
Data Structures | Enumerations | Functions
infer.h File Reference
#include "ove/types.h"
#include "ove_config.h"
#include "ove/storage.h"
Include dependency graph for infer.h:
This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures

struct  ove_tensor_info
 Tensor descriptor returned by ove_model_input() / ove_model_output(). More...
 
struct  ove_model_config
 Configuration for an ML inference session. More...
 

Enumerations

enum  ove_tensor_type {
  OVE_TENSOR_FLOAT32 = 0 , OVE_TENSOR_INT8 = 1 , OVE_TENSOR_UINT8 = 2 , OVE_TENSOR_INT16 = 3 ,
  OVE_TENSOR_INT32 = 4
}
 Tensor element types. More...
 

Functions

int ove_model_init (ove_model_t *model, ove_model_storage_t *storage, void *arena, const struct ove_model_config *cfg)
 Initialise a model using caller-supplied storage and arena.
 
void ove_model_deinit (ove_model_t model)
 Release resources held by a model initialised with ove_model_init().
 
int ove_model_create (ove_model_t *model, const struct ove_model_config *cfg)
 Allocate and initialise a model from the heap.
 
void ove_model_destroy (ove_model_t model)
 Destroy and free a model allocated with ove_model_create().
 
int ove_model_invoke (ove_model_t model)
 Run inference on the currently populated input tensor(s).
 
int ove_model_input (ove_model_t model, unsigned int index, struct ove_tensor_info *info)
 Get a descriptor for an input tensor.
 
int ove_model_output (ove_model_t model, unsigned int index, struct ove_tensor_info *info)
 Get a descriptor for an output tensor.
 
uint64_t ove_model_last_inference_us (ove_model_t model)
 Return inference time of the last ove_model_invoke() in microseconds.