ember.cpp 0.1.0
C++ inference engine for Llama-family and paperkiln models (CPU, CUDA, WebAssembly)
Loading...
Searching...
No Matches
llama3_tokenizer.h File Reference
#include <algorithm>
#include <cstdint>
#include <queue>
#include <stdexcept>
#include <string>
#include <unordered_map>
#include <utility>
#include <vector>
#include "bpe_tokenizer.h"
#include "llama3_unicode_tables.h"
Include dependency graph for llama3_tokenizer.h:
This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Classes

struct  microtorch::llama3::CptFlags
 
class  microtorch::llama3::Tokenizer
 
struct  microtorch::llama3::Tokenizer::Symbol
 
struct  microtorch::llama3::Tokenizer::Bigram
 
struct  microtorch::llama3::Tokenizer::Later
 

Namespaces

namespace  microtorch
 
namespace  microtorch::llama3
 

Functions

uint16_t microtorch::llama3::category_bits (uint32_t cpt)
 
CptFlags microtorch::llama3::flags_of (uint32_t cpt)
 
std::vector< uint32_t > microtorch::llama3::cpts_from_utf8 (const std::string &s)
 
void microtorch::llama3::append_utf8 (uint32_t cpt, std::string &out)
 
std::vector< std::string > microtorch::llama3::split_words (const std::string &text, size_t max_digits=3)
 
const std::vector< std::string > & microtorch::llama3::byte_to_unicode ()
 
std::string microtorch::llama3::byte_encode (const std::string &word)
 
Tokenizer microtorch::llama3::load_gguf (const std::string &path)