diff options
| author | Alexey Parfenov <zxed@alkatrazstudio.net> | 2024-02-11 13:43:31 +0000 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2024-02-11 15:43:31 +0200 |
| commit | a803333a4e6fc534c93afe90d741bc2388bdec87 (patch) | |
| tree | 0f4781f58f4391691823cdabaaf604afae333155 /common/sampling.h | |
| parent | 684780141a08200ec98eba3e982dbafd1d0b5000 (diff) | |
common : use enums for sampler types (#5418)
* common: use enums for sampler types
* Apply suggestions from code review
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
* minor : spaces
---------
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to 'common/sampling.h')
| -rw-r--r-- | common/sampling.h | 20 |
1 files changed, 19 insertions, 1 deletions
diff --git a/common/sampling.h b/common/sampling.h index 88899c09..2bd6a75d 100644 --- a/common/sampling.h +++ b/common/sampling.h @@ -8,6 +8,16 @@ #include <vector> #include <unordered_map> +// sampler types +enum class llama_sampler_type : char { + TOP_K = 'k', + TOP_P = 'p', + MIN_P = 'm', + TFS_Z = 'f', + TYPICAL_P = 'y', + TEMP = 't' +}; + // sampling parameters typedef struct llama_sampling_params { int32_t n_prev = 64; // number of previous tokens to remember @@ -28,7 +38,15 @@ typedef struct llama_sampling_params { float mirostat_tau = 5.00f; // target entropy float mirostat_eta = 0.10f; // learning rate bool penalize_nl = true; // consider newlines as a repeatable token - std::string samplers_sequence = "kfypmt"; // top_k, tail_free, typical_p, top_p, min_p, temp + + std::vector<llama_sampler_type> samplers_sequence = { + llama_sampler_type::TOP_K, + llama_sampler_type::TFS_Z, + llama_sampler_type::TYPICAL_P, + llama_sampler_type::TOP_P, + llama_sampler_type::MIN_P, + llama_sampler_type::TEMP + }; std::string grammar; // optional BNF-like grammar to constrain sampling |