ember.cpp 0.1.0
C++ inference engine for Llama-family and paperkiln models (CPU, CUDA, WebAssembly)
Loading...
Searching...
No Matches
ember_ir::Interpreter::Weight Struct Reference
Collaboration diagram for ember_ir::Interpreter::Weight:
Collaboration graph

Public Member Functions

bool quantised () const
 

Public Attributes

std::vector< float > f32
 
std::vector< uint8_t > q
 
const uint8_t * qp = nullptr
 
size_t qbytes = 0
 
GGMLType type = GGMLType::GGML_TYPE_F32
 
size_t numel = 0
 

Detailed Description

Definition at line 234 of file graph_ir.h.

Member Function Documentation

◆ quantised()

bool ember_ir::Interpreter::Weight::quantised ( ) const
inline

Definition at line 241 of file graph_ir.h.

241{ return qp != nullptr; }

References qp.

Member Data Documentation

◆ f32

std::vector<float> ember_ir::Interpreter::Weight::f32

Definition at line 235 of file graph_ir.h.

Referenced by ember_ir::Interpreter::bind_tensor().

◆ numel

size_t ember_ir::Interpreter::Weight::numel = 0

Definition at line 240 of file graph_ir.h.

Referenced by ember_ir::Interpreter::bind_tensor().

◆ q

std::vector<uint8_t> ember_ir::Interpreter::Weight::q

Definition at line 236 of file graph_ir.h.

◆ qbytes

size_t ember_ir::Interpreter::Weight::qbytes = 0

Definition at line 238 of file graph_ir.h.

◆ qp

const uint8_t* ember_ir::Interpreter::Weight::qp = nullptr

Definition at line 237 of file graph_ir.h.

Referenced by ember_ir::Interpreter::bind_tensor(), and quantised().

◆ type

GGMLType ember_ir::Interpreter::Weight::type = GGMLType::GGML_TYPE_F32

Definition at line 239 of file graph_ir.h.


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