diff options
Diffstat (limited to 'common/common.h')
| -rw-r--r-- | common/common.h | 88 |
1 files changed, 46 insertions, 42 deletions
diff --git a/common/common.h b/common/common.h index a8e5e50e..f68f3c29 100644 --- a/common/common.h +++ b/common/common.h @@ -27,7 +27,7 @@ #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__, LLAMA_BUILD_NUMBER, LLAMA_COMMIT); \ + fprintf(stderr, "%s: build = %d (%s)\n", __func__, LLAMA_BUILD_NUMBER, LLAMA_COMMIT); \ fprintf(stderr, "%s: built with %s for %s\n", __func__, LLAMA_COMPILER, LLAMA_BUILD_TARGET); \ } while(0) @@ -35,14 +35,18 @@ // build info extern int LLAMA_BUILD_NUMBER; -extern char const *LLAMA_COMMIT; -extern char const *LLAMA_COMPILER; -extern char const *LLAMA_BUILD_TARGET; +extern char const * LLAMA_COMMIT; +extern char const * LLAMA_COMPILER; +extern char const * LLAMA_BUILD_TARGET; struct llama_control_vector_load_info; -int get_math_cpu_count(); -int32_t get_num_physical_cores(); +// +// CPU utils +// + +int32_t cpu_get_num_physical_cores(); +int32_t cpu_get_num_math(); // // CLI argument parsing @@ -51,7 +55,7 @@ int32_t get_num_physical_cores(); struct gpt_params { uint32_t seed = LLAMA_DEFAULT_SEED; // RNG seed - int32_t n_threads = get_math_cpu_count(); + int32_t n_threads = cpu_get_num_math(); int32_t n_threads_draft = -1; int32_t n_threads_batch = -1; // number of threads to use for batch processing (-1 = use n_threads) int32_t n_threads_batch_draft = -1; @@ -179,33 +183,34 @@ struct gpt_params { void gpt_params_handle_model_default(gpt_params & params); -bool parse_kv_override(const char * data, std::vector<llama_model_kv_override> & overrides); - -bool gpt_params_parse_ex(int argc, char ** argv, gpt_params & params); - -bool gpt_params_parse(int argc, char ** argv, gpt_params & params); +bool gpt_params_parse_ex (int argc, char ** argv, gpt_params & params); +bool gpt_params_parse (int argc, char ** argv, gpt_params & params); +bool gpt_params_find_arg (int argc, char ** argv, const std::string & arg, gpt_params & params, int & i, bool & invalid_param); +void gpt_params_print_usage(int argc, char ** argv, const gpt_params & params); -void gpt_print_usage(int argc, char ** argv, const gpt_params & params); +std::string gpt_params_get_system_info(const gpt_params & params); -bool gpt_params_find_arg(int argc, char ** argv, const std::string & arg, gpt_params & params, int & i, bool & invalid_param); - -std::string get_system_info(const gpt_params & params); +// +// String utils +// -std::string gpt_random_prompt(std::mt19937 & rng); +std::vector<std::string> string_split(std::string input, char separator); -void process_escapes(std::string& input); +std::string string_strip(const std::string & str); +std::string string_get_sortable_timestamp(); +std::string string_random_prompt(std::mt19937 & rng); -bool validate_file_name(const std::string & filename); +bool string_parse_kv_override(const char * data, std::vector<llama_model_kv_override> & overrides); +void string_process_escapes(std::string & input); // -// String utils +// Filesystem utils // -std::vector<llama_sampler_type> sampler_types_from_names(const std::vector<std::string> & names, bool allow_alt_names); -std::vector<llama_sampler_type> sampler_types_from_chars(const std::string & names_string); -std::vector<std::string> string_split(std::string input, char separator); -std::string string_strip(const std::string & str); -std::string sampler_type_to_name_string(llama_sampler_type sampler_type); +bool fs_validate_filename(const std::string & filename); +bool fs_create_directory_with_parents(const std::string & path); + +std::string fs_get_cache_directory(); // // Model utils @@ -277,29 +282,14 @@ std::string llama_detokenize_bpe( bool llama_should_add_bos_token(const llama_model * model); // -// YAML utils -// - -bool create_directory_with_parents(const std::string & path); -std::string get_cache_directory(); -void dump_vector_float_yaml(FILE * stream, const char * prop_name, const std::vector<float> & data); -void dump_vector_int_yaml(FILE * stream, const char * prop_name, const std::vector<int> & data); -void dump_string_yaml_multiline(FILE * stream, const char * prop_name, const char * data); -std::string get_sortable_timestamp(); - -void dump_non_result_info_yaml( - FILE * stream, const gpt_params & params, const llama_context * lctx, - const std::string & timestamp, const std::vector<int> & prompt_tokens, const char * model_desc); - -// // KV cache utils // // Dump the KV cache view with the number of sequences per cell. -void dump_kv_cache_view(const llama_kv_cache_view & view, int row_size = 80); +void llama_kv_cache_dump_view(const llama_kv_cache_view & view, int row_size = 80); // Dump the KV cache view showing individual sequences in each cell (long output). -void dump_kv_cache_view_seqs(const llama_kv_cache_view & view, int row_size = 40); +void llama_kv_cache_dump_view_seqs(const llama_kv_cache_view & view, int row_size = 40); // // Embedding utils @@ -333,6 +323,20 @@ llama_control_vector_data llama_control_vector_load(const std::vector<llama_cont // // Split utils // + static const char * const LLM_KV_SPLIT_NO = "split.no"; static const char * const LLM_KV_SPLIT_COUNT = "split.count"; static const char * const LLM_KV_SPLIT_TENSORS_COUNT = "split.tensors.count"; + +// +// YAML utils +// + +void yaml_dump_vector_float (FILE * stream, const char * prop_name, const std::vector<float> & data); +void yaml_dump_vector_int (FILE * stream, const char * prop_name, const std::vector<int> & data); +void yaml_dump_string_multiline(FILE * stream, const char * prop_name, const char * data); + +void yaml_dump_non_result_info( + FILE * stream, const gpt_params & params, const llama_context * lctx, + const std::string & timestamp, const std::vector<int> & prompt_tokens, const char * model_desc); + |