8000 feat: Allow overriding GGUF metadata when loading model by KerfuffleV2 · Pull Request #4092 · ggml-org/llama.cpp · GitHub
[go: up one dir, main page]
Skip to content

feat: Allow overriding GGUF metadata when loading model #4092

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 13 commits into from
Dec 5, 2023
Merged

feat: Allow overriding GGUF metadata when loading model #4092

Changes from 1 commit
Commits
Show all changes
13 commits
Select commit Hold shift + click to select a range
ba839d1
feat: Allow overriding GGUF metadata when loading model
KerfuffleV2 Nov 15, 2023
9d39dea
Fix the one time GCC is stricter than clang about something
KerfuffleV2 Nov 15, 2023
69be5c3
Step1
KerfuffleV2 Nov 16, 2023
8c9f776
Refactor... basically everything!
KerfuffleV2 Nov 17, 2023
1844207
Nuke obsolete GetArrayLen struct
KerfuffleV2 Nov 17, 2023
dd89015
simplify std::string specialization
cebtenzzre Nov 17, 2023
cb5bfe0
Various cleanups
KerfuffleV2 Nov 18, 2023
2147421
Merge branch 'master' into feat-override-metadata
KerfuffleV2 Nov 18, 2023
aa7cf31
Fix broken logic for parsing bool KV overrides
KerfuffleV2 Nov 18, 2023
a5b7d72
Merge branch 'master' into feat-override-metadata
KerfuffleV2 Nov 24, 2023
14e0ba1
llama : rearrange model params
ggerganov Dec 5, 2023 8000
a59bf93
Merge branch 'master' into feat-override-metadata
KerfuffleV2 Dec 5, 2023
7bbe605
Update new GET_KEY call
KerfuffleV2 Dec 5, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Prev Previous commit
Next Next commit
Fix the one time GCC is stricter than clang about something
  • Loading branch information
@KerfuffleV2
KerfuffleV2 committed Nov 15, 2023
commit 9d39deab8fea3b3b0a1430768c7903a3170a9bcc
24 changes: 12 additions & 12 deletions llama.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1835,18 +1835,18 @@ struct llama_model_loader {
throw std::runtime_error(format("request for key id %d with unhandled result type: %s", kid, typeid(T).name()));
}

template<> void gk_set(int k, uint8_t & r) { r = gguf_get_val_u8 (ctx_gguf, k); }
template<> void gk_set(int k, uint16_t & r) { r = gguf_get_val_u16 (ctx_gguf, k); }
template<> void gk_set(int k, uint32_t & r) { r = gguf_get_val_u32 (ctx_gguf, k); }
template<> void gk_set(int k, uint64_t & r) { r = gguf_get_val_u64 (ctx_gguf, k); }
template<> void gk_set(int k, int8_t & r) { r = gguf_get_val_i8 (ctx_gguf, k); }
template<> void gk_set(int k, int16_t & r) { r = gguf_get_val_i16 (ctx_gguf, k); }
template<> void gk_set(int k, int32_t & r) { r = gguf_get_val_i32 (ctx_gguf, k); }
template<> void gk_set(int k, int64_t & r) { r = gguf_get_val_i64 (ctx_gguf, k); }
template<> void gk_set(int k, float & r) { r = gguf_get_val_f32 (ctx_gguf, k); }
template<> void gk_set(int k, double & r) { r = gguf_get_val_f64 (ctx_gguf, k); }
template<> void gk_set(int k, bool & r) { r = gguf_get_val_bool(ctx_gguf, k); }
template<> void gk_set(int k, std::string & r) { r = std::string(gguf_get_val_str(ctx_gguf, k)); }
void gk_set(int k, uint8_t & r) { r = gguf_get_val_u8 (ctx_gguf, k); }
void gk_set(int k, uint16_t & r) { r = gguf_get_val_u16 (ctx_gguf, k); }
void gk_set(int k, uint32_t & r) { r = gguf_get_val_u32 (ctx_gguf, k); }
void gk_set(int k, uint64_t & r) { r = gguf_get_val_u64 (ctx_gguf, k); }
void gk_set(int k, int8_t & r) { r = gguf_get_val_i8 (ctx_gguf, k); }
void gk_set(int k, int16_t & r) { r = gguf_get_val_i16 (ctx_gguf, k); }
void gk_set(int k, int32_t & r) { r = gguf_get_val_i32 (ctx_gguf, k); }
void gk_set(int k, int64_t & r) { r = gguf_get_val_i64 (ctx_gguf, k); }
void gk_set(int k, float & r) { r = gguf_get_val_f32 (ctx_gguf, k); }
void gk_set(int k, double & r) { r = gguf_get_val_f64 (ctx_gguf, k); }
void gk_set(int k, bool & r) { r = gguf_get_val_bool(ctx_gguf, k); }
void gk_set(int k, std::string & r) { r = std::string(gguf_get_val_str(ctx_gguf, k)); }

template<typename T>
typename std::enable_if<std::is_integral<T>::value, void>::type
Expand Down
0