Merge branch 'master' into custom-attention-mask

This commit is contained in:
Georgi Gerganov 2023-09-18 11:15:18 +03:00
commit 58bb5110ca
No known key found for this signature in database
GPG key ID: 449E073F9DC10735
44 changed files with 1180 additions and 413 deletions

View file

@ -78,7 +78,7 @@ int32_t get_num_physical_cores() {
return n_threads > 0 ? (n_threads <= 4 ? n_threads : n_threads / 2) : 4;
}
void process_escapes(std::string& input) {
static void process_escapes(std::string& input) {
std::size_t input_len = input.length();
std::size_t output_idx = 0;
@ -798,10 +798,10 @@ std::vector<llama_token> llama_tokenize(
// upper limit for the number of tokens
int n_tokens = text.length() + add_bos;
std::vector<llama_token> result(n_tokens);
n_tokens = llama_tokenize(ctx, text.c_str(), result.data(), result.size(), add_bos);
n_tokens = llama_tokenize(ctx, text.data(), text.length(), result.data(), result.size(), add_bos);
if (n_tokens < 0) {
result.resize(-n_tokens);
int check = llama_tokenize(ctx, text.c_str(), result.data(), result.size(), add_bos);
int check = llama_tokenize(ctx, text.data(), text.length(), result.data(), result.size(), add_bos);
GGML_ASSERT(check == -n_tokens);
} else {
result.resize(n_tokens);

View file

@ -3,6 +3,7 @@
#pragma once
#include "llama.h"
#include "build-info.h"
#define LOG_NO_FILE_LINE_FUNCTION
#include "log.h"
@ -23,6 +24,11 @@
#define die(msg) do { fputs("error: " msg "\n", stderr); exit(1); } while (0)
#define die_fmt(fmt, ...) do { fprintf(stderr, "error: " fmt "\n", __VA_ARGS__); exit(1); } while (0)
#define print_build_info() do { \
fprintf(stderr, "%s: build = %d (%s)\n", __func__, BUILD_NUMBER, BUILD_COMMIT); \
fprintf(stderr, "%s: built with %s for %s\n", __func__, BUILD_COMPILER, BUILD_TARGET); \
} while(0)
//
// CLI argument parsing
//

View file

@ -158,7 +158,7 @@ namespace console {
}
}
char32_t getchar32() {
static char32_t getchar32() {
#if defined(_WIN32)
HANDLE hConsole = GetStdHandle(STD_INPUT_HANDLE);
wchar_t high_surrogate = 0;
@ -212,7 +212,7 @@ namespace console {
#endif
}
void pop_cursor() {
static void pop_cursor() {
#if defined(_WIN32)
if (hConsole != NULL) {
CONSOLE_SCREEN_BUFFER_INFO bufferInfo;
@ -233,7 +233,7 @@ namespace console {
putc('\b', out);
}
int estimateWidth(char32_t codepoint) {
static int estimateWidth(char32_t codepoint) {
#if defined(_WIN32)
(void)codepoint;
return 1;
@ -242,7 +242,7 @@ namespace console {
#endif
}
int put_codepoint(const char* utf8_codepoint, size_t length, int expectedWidth) {
static int put_codepoint(const char* utf8_codepoint, size_t length, int expectedWidth) {
#if defined(_WIN32)
CONSOLE_SCREEN_BUFFER_INFO bufferInfo;
if (!GetConsoleScreenBufferInfo(hConsole, &bufferInfo)) {
@ -303,7 +303,7 @@ namespace console {
#endif
}
void replace_last(char ch) {
static void replace_last(char ch) {
#if defined(_WIN32)
pop_cursor();
put_codepoint(&ch, 1, 1);
@ -312,7 +312,7 @@ namespace console {
#endif
}
void append_utf8(char32_t ch, std::string & out) {
static void append_utf8(char32_t ch, std::string & out) {
if (ch <= 0x7F) {
out.push_back(static_cast<unsigned char>(ch));
} else if (ch <= 0x7FF) {
@ -333,7 +333,7 @@ namespace console {
}
// Helper function to remove the last UTF-8 character from a string
void pop_back_utf8_char(std::string & line) {
static void pop_back_utf8_char(std::string & line) {
if (line.empty()) {
return;
}
@ -349,7 +349,7 @@ namespace console {
line.erase(pos);
}
bool readline_advanced(std::string & line, bool multiline_input) {
static bool readline_advanced(std::string & line, bool multiline_input) {
if (out != stdout) {
fflush(stdout);
}
@ -452,7 +452,7 @@ namespace console {
return has_more;
}
bool readline_simple(std::string & line, bool multiline_input) {
static bool readline_simple(std::string & line, bool multiline_input) {
#if defined(_WIN32)
std::wstring wline;
if (!std::getline(std::wcin, wline)) {

View file

@ -9,7 +9,7 @@
namespace grammar_parser {
// NOTE: assumes valid utf8 (but checks for overrun)
// copied from llama.cpp
std::pair<uint32_t, const char *> decode_utf8(const char * src) {
static std::pair<uint32_t, const char *> decode_utf8(const char * src) {
static const int lookup[] = { 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 2, 2, 3, 4 };
uint8_t first_byte = static_cast<uint8_t>(*src);
uint8_t highbits = first_byte >> 4;
@ -24,19 +24,19 @@ namespace grammar_parser {
return std::make_pair(value, pos);
}
uint32_t get_symbol_id(parse_state & state, const char * src, size_t len) {
static uint32_t get_symbol_id(parse_state & state, const char * src, size_t len) {
uint32_t next_id = static_cast<uint32_t>(state.symbol_ids.size());
auto result = state.symbol_ids.insert(std::make_pair(std::string(src, len), next_id));
return result.first->second;
}
uint32_t generate_symbol_id(parse_state & state, const std::string & base_name) {
static uint32_t generate_symbol_id(parse_state & state, const std::string & base_name) {
uint32_t next_id = static_cast<uint32_t>(state.symbol_ids.size());
state.symbol_ids[base_name + '_' + std::to_string(next_id)] = next_id;
return next_id;
}
void add_rule(
static void add_rule(
parse_state & state,
uint32_t rule_id,
const std::vector<llama_grammar_element> & rule) {
@ -46,11 +46,11 @@ namespace grammar_parser {
state.rules[rule_id] = rule;
}
bool is_word_char(char c) {
static bool is_word_char(char c) {
return ('a' <= c && c <= 'z') || ('A' <= c && c <= 'Z') || c == '-' || ('0' <= c && c <= '9');
}
std::pair<uint32_t, const char *> parse_hex(const char * src, int size) {
static std::pair<uint32_t, const char *> parse_hex(const char * src, int size) {
const char * pos = src;
const char * end = src + size;
uint32_t value = 0;
@ -73,7 +73,7 @@ namespace grammar_parser {
return std::make_pair(value, pos);
}
const char * parse_space(const char * src, bool newline_ok) {
static const char * parse_space(const char * src, bool newline_ok) {
const char * pos = src;
while (*pos == ' ' || *pos == '\t' || *pos == '#' ||
(newline_ok && (*pos == '\r' || *pos == '\n'))) {
@ -88,7 +88,7 @@ namespace grammar_parser {
return pos;
}
const char * parse_name(const char * src) {
static const char * parse_name(const char * src) {
const char * pos = src;
while (is_word_char(*pos)) {
pos++;
@ -99,7 +99,7 @@ namespace grammar_parser {
return pos;
}
std::pair<uint32_t, const char *> parse_char(const char * src) {
static std::pair<uint32_t, const char *> parse_char(const char * src) {
if (*src == '\\') {
switch (src[1]) {
case 'x': return parse_hex(src + 2, 2);
@ -129,7 +129,7 @@ namespace grammar_parser {
uint32_t rule_id,
bool is_nested);
const char * parse_sequence(
static const char * parse_sequence(
parse_state & state,
const char * src,
const std::string & rule_name,
@ -247,7 +247,7 @@ namespace grammar_parser {
return pos;
}
const char * parse_rule(parse_state & state, const char * src) {
static const char * parse_rule(parse_state & state, const char * src) {
const char * name_end = parse_name(src);
const char * pos = parse_space(name_end, false);
size_t name_len = name_end - src;
@ -285,7 +285,7 @@ namespace grammar_parser {
}
}
void print_grammar_char(FILE * file, uint32_t c) {
static void print_grammar_char(FILE * file, uint32_t c) {
if (0x20 <= c && c <= 0x7f) {
fprintf(file, "%c", static_cast<char>(c));
} else {
@ -294,7 +294,7 @@ namespace grammar_parser {
}
}
bool is_char_element(llama_grammar_element elem) {
static bool is_char_element(llama_grammar_element elem) {
switch (elem.type) {
case LLAMA_GRETYPE_CHAR: return true;
case LLAMA_GRETYPE_CHAR_NOT: return true;
@ -304,7 +304,7 @@ namespace grammar_parser {
}
}
void print_rule_binary(FILE * file, const std::vector<llama_grammar_element> & rule) {
static void print_rule_binary(FILE * file, const std::vector<llama_grammar_element> & rule) {
for (auto elem : rule) {
switch (elem.type) {
case LLAMA_GRETYPE_END: fprintf(file, "END"); break;
@ -334,7 +334,7 @@ namespace grammar_parser {
fprintf(file, "\n");
}
void print_rule(
static void print_rule(
FILE * file,
uint32_t rule_id,
const std::vector<llama_grammar_element> & rule,