Oakfield API Documentation 1.0.0
Numerical core APIs
Loading...
Searching...
No Matches
SimNeuralTensorMapping Struct Reference

Opaque-ish mapping handle used by input/output mapping paths. More...

#include <neural_tensor_map.h>

Collaboration diagram for SimNeuralTensorMapping:

Public Attributes

SimNeuralTensorView view
 
SimNeuralTensorMapMetrics metrics
 
SimNeuralTensorMapConfig config
 
const SimFieldfield_const
 
SimFieldfield_mut
 
void * scratch
 
size_t scratch_bytes
 
uint8_t field_rank
 
uint8_t field_channel_axis
 
uint8_t field_spatial_rank
 
uint8_t field_spatial_axes [3]
 
bool has_field_channel_axis
 
bool output_mode
 
bool active
 

Detailed Description

Opaque-ish mapping handle used by input/output mapping paths.

Call sim_neural_tensor_mapping_release after use. For output mappings created by sim_neural_tensor_map_output, call sim_neural_tensor_unmap_output before release when fallback copy was used.

Member Data Documentation

◆ active

bool SimNeuralTensorMapping::active

True while the mapping owns/references live resources.

◆ config

SimNeuralTensorMapConfig SimNeuralTensorMapping::config

Normalized mapping policy.

◆ field_channel_axis

uint8_t SimNeuralTensorMapping::field_channel_axis

Resolved channel axis in field coordinates.

◆ field_const

const SimField* SimNeuralTensorMapping::field_const

Borrowed source field for input mappings.

◆ field_mut

SimField* SimNeuralTensorMapping::field_mut

Borrowed mutable field for output mappings.

◆ field_rank

uint8_t SimNeuralTensorMapping::field_rank

Rank of the mapped field layout.

◆ field_spatial_axes

uint8_t SimNeuralTensorMapping::field_spatial_axes[3]

Field axes corresponding to tensor spatial dimensions.

◆ field_spatial_rank

uint8_t SimNeuralTensorMapping::field_spatial_rank

Number of spatial axes in the mapped field.

◆ has_field_channel_axis

bool SimNeuralTensorMapping::has_field_channel_axis

True when the field exposes an explicit channel axis.

◆ metrics

SimNeuralTensorMapMetrics SimNeuralTensorMapping::metrics

Copy/transform costs observed during mapping.

◆ output_mode

bool SimNeuralTensorMapping::output_mode

True when mapping is intended for write-back.

◆ scratch

void* SimNeuralTensorMapping::scratch

Owned fallback tensor buffer, released by mapping_release().

◆ scratch_bytes

size_t SimNeuralTensorMapping::scratch_bytes

Allocated byte size of scratch.

◆ view

SimNeuralTensorView SimNeuralTensorMapping::view

Tensor view exposed to the backend.


The documentation for this struct was generated from the following file: