ExternalVendorCode/whisper.cpp
1
0
Fork 0
mirror of https://github.com/ggerganov/whisper.cpp.git synced 2025-06-13 04:28:07 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

talk-llama : add --session support (#845)

Browse Source 
* feat: adding session support

* readme: adding --session info in examples/talk-llama

* llama: adding session fixes

* readme: updating session doc

* talk-llama: update the value of need_to_save_session to true in order to save the session in the subsequent interaction

* talk-llama: adding missing function which updates session_tokens
This commit is contained in:
Luis Herrera
2023-05-01 12:18:10 -05:00
committed by GitHub
parent d375d73b2e
commit be5911a9f3
4 changed files with 171 additions and 42 deletions
Download Patch File Download Diff File Expand all files Collapse all files

13
examples/talk-llama/llama.h
View File

@ -19,9 +19,11 @@
# define LLAMA_API
#endif
#define LLAMA_FILE_VERSION 1
#define LLAMA_FILE_MAGIC 0x67676a74 // 'ggjt' in hex
#define LLAMA_FILE_MAGIC_UNVERSIONED 0x67676d6c // pre-versioned files
#define LLAMA_FILE_VERSION 1
#define LLAMA_FILE_MAGIC 'ggjt'
#define LLAMA_FILE_MAGIC_UNVERSIONED 'ggml'
#define LLAMA_SESSION_MAGIC 'ggsn'
#define LLAMA_SESSION_VERSION 0
#ifdef __cplusplus
extern "C" {
@ -138,9 +140,8 @@ extern "C" {
LLAMA_API size_t llama_set_state_data(struct llama_context * ctx, const uint8_t * src);
// Save/load session file
LLAMA_API size_t llama_load_session_file(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out);
LLAMA_API size_t llama_save_session_file(struct llama_context * ctx, const char * path_session, const llama_token * tokens, size_t n_token_count);
LLAMA_API bool llama_load_session_file(struct llama_context * ctx, const char * path_session, llama_token * tokens_out, size_t n_token_capacity, size_t * n_token_count_out);
LLAMA_API bool llama_save_session_file(struct llama_context * ctx, const char * path_session, const llama_token * tokens, size_t n_token_count);
// Run the llama inference to obtain the logits and probabilities for the next token.
// tokens + n_tokens is the provided batch of new tokens to process
// n_past is the number of tokens to use from previous eval calls