From 154e0d75fccf1784fe9ff6fd76a630b66563da3d Mon Sep 17 00:00:00 2001 From: Kawrakow <48489457+ikawrakow@users.noreply.github.com> Date: Sat, 27 Jul 2024 07:55:01 +0200 Subject: Merge mainline llama.cpp (#3) * Merging mainline - WIP * Merging mainline - WIP AVX2 and CUDA appear to work. CUDA performance seems slightly (~1-2%) lower as it is so often the case with llama.cpp/ggml after some "improvements" have been made. * Merging mainline - fix Metal * Remove check --------- Co-authored-by: Iwan Kawrakow --- unicode.h | 63 --------------------------------------------------------------- 1 file changed, 63 deletions(-) delete mode 100644 unicode.h (limited to 'unicode.h') diff --git a/unicode.h b/unicode.h deleted file mode 100644 index 6c488970..00000000 --- a/unicode.h +++ /dev/null @@ -1,63 +0,0 @@ -#pragma once - -#include -#include -#include - -struct codepoint_flags { - enum { - UNDEFINED = 0x0001, - NUMBER = 0x0002, // regex: \p{N} - LETTER = 0x0004, // regex: \p{L} - SEPARATOR = 0x0008, // regex: \p{Z} - ACCENT_MARK = 0x0010, // regex: \p{M} - PUNCTUATION = 0x0020, // regex: \p{P} - SYMBOL = 0x0040, // regex: \p{S} - CONTROL = 0x0080, // regex: \p{C} - MASK_CATEGORIES = 0x00FF, - }; - - // codepoint type - uint16_t is_undefined : 1; - uint16_t is_number : 1; // regex: \p{N} - uint16_t is_letter : 1; // regex: \p{L} - uint16_t is_separator : 1; // regex: \p{Z} - uint16_t is_accent_mark : 1; // regex: \p{M} - uint16_t is_punctuation : 1; // regex: \p{P} - uint16_t is_symbol : 1; // regex: \p{S} - uint16_t is_control : 1; // regex: \p{C} - // helper flags - uint16_t is_whitespace : 1; // regex: \s - uint16_t is_lowercase : 1; - uint16_t is_uppercase : 1; - uint16_t is_nfd : 1; - - // decode from uint16 - inline codepoint_flags(const uint16_t flags=0) { - *reinterpret_cast(this) = flags; - } - - inline uint16_t as_uint() const { - return *reinterpret_cast(this); - } - - inline uint16_t category_flag() const { - return this->as_uint() & MASK_CATEGORIES; - } -}; - - -std::string unicode_cpt_to_utf8(uint32_t cp); -std::vector unicode_cpts_from_utf8(const std::string & utf8); - -std::vector unicode_cpts_normalize_nfd(const std::vector & cpts); - -codepoint_flags unicode_cpt_flags(const uint32_t cp); -codepoint_flags unicode_cpt_flags(const std::string & utf8); - -std::string unicode_byte_to_utf8(uint8_t byte); -uint8_t unicode_utf8_to_byte(const std::string & utf8); - -uint32_t unicode_tolower(uint32_t cp); - -std::vector unicode_regex_split(const std::string & text, const std::vector & regex_exprs); -- cgit v1.2.3