spec : update CLI arguments for better consistency (#22964)
* spec : update CLI arguments for better consistency * cont : fix CLI arg message
This commit is contained in:
+4
-3
@@ -157,9 +157,9 @@ enum common_params_sampling_config : uint64_t {
|
||||
|
||||
enum common_speculative_type {
|
||||
COMMON_SPECULATIVE_TYPE_NONE, // no speculative decoding
|
||||
COMMON_SPECULATIVE_TYPE_DRAFT, // draft model
|
||||
COMMON_SPECULATIVE_TYPE_EAGLE3, // eagle draft model
|
||||
COMMON_SPECULATIVE_TYPE_NGRAM_SIMPLE, // simple self-speculative decoding
|
||||
COMMON_SPECULATIVE_TYPE_DRAFT_SIMPLE, // standalone draft model speculative decoding
|
||||
COMMON_SPECULATIVE_TYPE_DRAFT_EAGLE3, // Eagle3 speculative decoding
|
||||
COMMON_SPECULATIVE_TYPE_NGRAM_SIMPLE, // simple self-speculative decoding based on n-grams
|
||||
COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K, // self-speculative decoding with n-gram keys only
|
||||
COMMON_SPECULATIVE_TYPE_NGRAM_MAP_K4V, // self-speculative decoding with n-gram keys and 4 m-gram values
|
||||
COMMON_SPECULATIVE_TYPE_NGRAM_MOD,
|
||||
@@ -342,6 +342,7 @@ struct common_params_speculative_ngram_cache {
|
||||
struct common_params_speculative {
|
||||
std::vector<enum common_speculative_type> types = { COMMON_SPECULATIVE_TYPE_NONE };
|
||||
|
||||
// used by Simple, MTP, Eagle3, etc. - all methods that require some kind of draft model
|
||||
common_params_speculative_draft draft;
|
||||
|
||||
common_params_speculative_ngram_mod ngram_mod;
|
||||
|
||||
Reference in New Issue
Block a user